Hi,
I didn't searched the whole forum for my question, because it seems to be impossible.... I ask is simply, perhaps somebody has any experience.
I compared two amps. One is an older, discrete, cheap commercial amp. The other is a composite diy IC amp. I've made some A-B listening test with them.
There are quality differences in the midrange and high section. It is plausible, because of the one has electrolytic capacitors in the signal path and cheap ceramics for compensating, the other has a DC servo and foil caps.
But what I do not understand is the low-mid and bass section. One has more punch, detail, double bass and chello e.g. sounds with more body and belivable "resonance". The other sounds more "flat", without dinamic, little boring.
Why could be these difference by the lows? I know, there are completly different circuits, but theoretically what determines the dynamic and details of the low-mid / bass sections?
I didn't searched the whole forum for my question, because it seems to be impossible.... I ask is simply, perhaps somebody has any experience.
I compared two amps. One is an older, discrete, cheap commercial amp. The other is a composite diy IC amp. I've made some A-B listening test with them.
There are quality differences in the midrange and high section. It is plausible, because of the one has electrolytic capacitors in the signal path and cheap ceramics for compensating, the other has a DC servo and foil caps.
But what I do not understand is the low-mid and bass section. One has more punch, detail, double bass and chello e.g. sounds with more body and belivable "resonance". The other sounds more "flat", without dinamic, little boring.
Why could be these difference by the lows? I know, there are completly different circuits, but theoretically what determines the dynamic and details of the low-mid / bass sections?
Last edited:
I would be doing some sweeps with a signal generator and a scope on the output to see what the differences are.
Perhaps the older one is getting near needing recapping ?
I cant see DC servo making any difference as that is very low frequency and pretty much DC.
Perhaps the older one is getting near needing recapping ?
I cant see DC servo making any difference as that is very low frequency and pretty much DC.
It's intresting, but the older amp has the better low-freq sound. Old 2x6.800uF "Marcon" caps. The newer has fresh 2x10.000uF Vishay BC 056 and MUR-820 diodes...
Last edited:
The analytical and the intuitive...
There are many parts in the stream of the sound, and much can do harm. How 'well tempered' is a design, the building of it, the reliability in due time?
How do we perceive? (Nice topic in the Lounge section: Limits of Audibility). What can we perceive?
The lower frequencies are sensed in the phase region, whereas the higher frequencies are more amplitude related, due to the evolution of the hearing devices of mammals we're part of. The high pitched rustling reeds does not invoke uneasyness, where the distant deep thumb of the hungry lion's claw in the earth stirrs our awareness immediately.
There are many parts in the stream of the sound, and much can do harm. How 'well tempered' is a design, the building of it, the reliability in due time?
How do we perceive? (Nice topic in the Lounge section: Limits of Audibility). What can we perceive?
The lower frequencies are sensed in the phase region, whereas the higher frequencies are more amplitude related, due to the evolution of the hearing devices of mammals we're part of. The high pitched rustling reeds does not invoke uneasyness, where the distant deep thumb of the hungry lion's claw in the earth stirrs our awareness immediately.
There’s far more to the PSU than just the caps. Wire gauge / track thickness and the rating of the transformers, for example. IMHO the PSU is the most important part of an amplifier.
Paul
Paul
In any such comparison, the things I would worry about are the absolute basics:
1. Do both amplifiers have a flat frequency response?
A measurement adapter built with a topology like this:
could be used to connect the microphone input of a typical USB audio interface, like a Focusrite Saffire or a MOTU M2. (You could try R1 = R3 = 22k, R2 = R4 = 470 ohm. 1/4W metal film.) Then you could use RMAA for loopback testing, or maybe REW. You might use the headphone output on the commercial amp for getting familiar with the process and getting first results.
2. Do both amplifiers have low output impedance (<160 mOhms or so)? Protection relay contacts in the older commercial amp may have tarnished or seen arcing, and may at the very least require a polish or a new relay altogether. For testing, you can try looking up the relay pinout and bridging the contacts that it connects in use with some wire. If there is a mechanical speaker selector switch, this may be problematic for similar reasons.
High output impedance would cause frequency response to follow the speaker impedance curve, often resulting in an effect as described.
3. Are both amplifiers precisely matched in output level (ideally to <0.1 dB), with direct A/B switching? That's quite possibly easier said than done, given the channel imbalance typically found in volume pots. An effort must definitely be made in this regard, as our perception of sound is substantially affected by level - basically, "louder = better". Also, our auditory memory is quite short (think seconds).
4. Do both amplifiers have negligible distortion throughout the audio band with a load connected? (Again, the loopback testing solution from 1. could be used.) Having a dummy load would be best but I guess a real speaker may be when using RMAA's speaker testing mode or at moderate levels (maybe 85 dB SPL or so).
1. Do both amplifiers have a flat frequency response?
A measurement adapter built with a topology like this:
could be used to connect the microphone input of a typical USB audio interface, like a Focusrite Saffire or a MOTU M2. (You could try R1 = R3 = 22k, R2 = R4 = 470 ohm. 1/4W metal film.) Then you could use RMAA for loopback testing, or maybe REW. You might use the headphone output on the commercial amp for getting familiar with the process and getting first results.
2. Do both amplifiers have low output impedance (<160 mOhms or so)? Protection relay contacts in the older commercial amp may have tarnished or seen arcing, and may at the very least require a polish or a new relay altogether. For testing, you can try looking up the relay pinout and bridging the contacts that it connects in use with some wire. If there is a mechanical speaker selector switch, this may be problematic for similar reasons.
High output impedance would cause frequency response to follow the speaker impedance curve, often resulting in an effect as described.
3. Are both amplifiers precisely matched in output level (ideally to <0.1 dB), with direct A/B switching? That's quite possibly easier said than done, given the channel imbalance typically found in volume pots. An effort must definitely be made in this regard, as our perception of sound is substantially affected by level - basically, "louder = better". Also, our auditory memory is quite short (think seconds).
4. Do both amplifiers have negligible distortion throughout the audio band with a load connected? (Again, the loopback testing solution from 1. could be used.) Having a dummy load would be best but I guess a real speaker may be when using RMAA's speaker testing mode or at moderate levels (maybe 85 dB SPL or so).
Attachments
Last edited:
I give you 100% credibility. The split power supply has halve wave rectified current of the output power. The best sounding power supply I achieved is the Single Floating Supply, used in high end Onkyo amps.
The wires/tracks should obey the AWG recommended thickness for power transmission and not it's copper resistance as usually applied.
The opening post contains the first problem in formulating a reply: How do we sensibly compare old with new and very different devices, technologies, components, circuit layout and so forth? The next is to question why available technical reviews and test data published by, for example Stereophile magazine, are not used to learn from and refer to. There is a wealth of information on the subjective and technical audio quality of commercial products out there - why not read some from reputable sources and gain a little more than a discussion without facts and figures can provide?
I find a lot differences between even the same type/model amplifiers before they are brought to a similar condition by say, a full service to an ancient relic of the 1970s. It should be clear enough to anyone aware of the deterioration of electronic components, switches, circuit boards etc, some perhaps with only a 10 year product lifetime, perhaps as the manufacturers intended.
When chipamps, SMT construction and fundamentally different technologies like class D operation are included in comparisons, the field is wide open since each technology has its differences and limitations. Some are consistent and characteristic of the design techniques and circuits used, others just reflect the limited options available to designers working to tight (often ridiculously low) budgets.
A low budget audio amplifier or other low cost device in the signal chain will likely have an adverse impact on the sound of a high-end system, however small. When every functional unit of the system is replaced by cheapest possible design and build quality types, I would expect to hear quite obvious differences. Otherwise, the need and desire for expensive gear would only be justified by aesthetics like form, feel and finish.
I find a lot differences between even the same type/model amplifiers before they are brought to a similar condition by say, a full service to an ancient relic of the 1970s. It should be clear enough to anyone aware of the deterioration of electronic components, switches, circuit boards etc, some perhaps with only a 10 year product lifetime, perhaps as the manufacturers intended.
When chipamps, SMT construction and fundamentally different technologies like class D operation are included in comparisons, the field is wide open since each technology has its differences and limitations. Some are consistent and characteristic of the design techniques and circuits used, others just reflect the limited options available to designers working to tight (often ridiculously low) budgets.
A low budget audio amplifier or other low cost device in the signal chain will likely have an adverse impact on the sound of a high-end system, however small. When every functional unit of the system is replaced by cheapest possible design and build quality types, I would expect to hear quite obvious differences. Otherwise, the need and desire for expensive gear would only be justified by aesthetics like form, feel and finish.
There are some possibilities, such as what the LF time constants are between the two amplifiers. I take it that the older design is direct coupled, and the two candidates are the feedback decoupling and input. These may be "adding something" to the music if not low enough.
But I think the only real way to investigate this is to measure the response. If the newer design has an LF servo it maybe that it is not low enough and not as close to DC as it should be.
It could be that the capacitors used have got some distortion too, especially if they are old. That would require some dedicated kit to measure.
And as another responder posted, it is important when making subjective comparisons to keep the actual volumes the same because of the perceived loudness and bass reponse being related.
In summary, all of the above!
But I think the only real way to investigate this is to measure the response. If the newer design has an LF servo it maybe that it is not low enough and not as close to DC as it should be.
It could be that the capacitors used have got some distortion too, especially if they are old. That would require some dedicated kit to measure.
And as another responder posted, it is important when making subjective comparisons to keep the actual volumes the same because of the perceived loudness and bass reponse being related.
In summary, all of the above!