I have an old receiver that produces some of the nicest bass and midrange I've encountered, though not the ultimate in accuracy. The sense of almost endless, rumbling low-frequency extension and smooth, relaxed midrange made far superior gear sound thin or uninteresting by comparison. I have a $4000 amplifier that doesn't even get use because of it, and others sold off.
I could never replicate the effect using a traditional digital parametric equalizer, so I'm entertaining a possibility that the receiver could behave like a compressor as a dynamic means of frequency response shaping.
I captured some casual measurements with REW. The levels of these measurements were normalized for comparison, since they couldn't be taken at the exact same levels. Measuring the amp output required a voltage divider network, while measuring the preamp output was done straight to a mic line in, so the levels were different prior to normalizing them.
In the first graph, the power amp section, pre amp section and the two together were tested between 0.5W and 1W, normal listening. Nothing special about the preamp section's response. I wonder if a condition, such as the input impedance of the power amp section was better suited to the preamp's output? Viewing the violet curve, the preamp has more extension when it's connected to the power amp section, but I can't compare levels to see if the response may have been compressed to extend it:
In the second graph, the measurements were also taken at slightly different levels and then normalized for comparison. We see the baseline amplifier reponse again (green), then the amplifier was operated with a real 3-way speaker load (brown). The response follows the speaker's impedance curve, due to the amp section's higher output impedance. It's also like something occurs between the pre and amplifier. I just don't know how to analyze a signal for compression.
Has anyone encountered a low-frequency enhancing behavior like this?
One wouldn't know it rolled off in the bass and would believe there was a 9dB tilt in response, though it's only 4dB as measured.
I could never replicate the effect using a traditional digital parametric equalizer, so I'm entertaining a possibility that the receiver could behave like a compressor as a dynamic means of frequency response shaping.
I captured some casual measurements with REW. The levels of these measurements were normalized for comparison, since they couldn't be taken at the exact same levels. Measuring the amp output required a voltage divider network, while measuring the preamp output was done straight to a mic line in, so the levels were different prior to normalizing them.
In the first graph, the power amp section, pre amp section and the two together were tested between 0.5W and 1W, normal listening. Nothing special about the preamp section's response. I wonder if a condition, such as the input impedance of the power amp section was better suited to the preamp's output? Viewing the violet curve, the preamp has more extension when it's connected to the power amp section, but I can't compare levels to see if the response may have been compressed to extend it:
In the second graph, the measurements were also taken at slightly different levels and then normalized for comparison. We see the baseline amplifier reponse again (green), then the amplifier was operated with a real 3-way speaker load (brown). The response follows the speaker's impedance curve, due to the amp section's higher output impedance. It's also like something occurs between the pre and amplifier. I just don't know how to analyze a signal for compression.
Has anyone encountered a low-frequency enhancing behavior like this?
One wouldn't know it rolled off in the bass and would believe there was a 9dB tilt in response, though it's only 4dB as measured.
Last edited:
Is the brand and model of this vintage receiver a secret? Otherwise, analysis of what was intended is only possible if we know the amplifier's relevant details including specs, schematic, maybe some evidence of repairs or tinkering and perhaps a published review reference or two?
Nonlinearities/intermodulation can help you perceive deep Bass.
You are not actually perceiving low frequencies but their influence on the rest of the program.
You are not actually perceiving low frequencies but their influence on the rest of the program.
Hi JMFahey,
I'm intrigued to try and uncover which type it is. Woofers do move, though, far more than with my other amps and past amps. I took some recordings of it driving a speaker in an anechoic chamber, and it's distinct.
Hi Ian Finch,
Brand and model: Nikko STA-8080. The service manual with schematic is available at vintageshifi.com
I'm intrigued to try and uncover which type it is. Woofers do move, though, far more than with my other amps and past amps. I took some recordings of it driving a speaker in an anechoic chamber, and it's distinct.
Hi Ian Finch,
Brand and model: Nikko STA-8080. The service manual with schematic is available at vintageshifi.com
Where does 4dB come from? The tilt from the graph is under 2dB although this will contribute to enhanced bass. The bump at 400Hz will also make bass warmer.
The output impedance of the preamp at LF seems high. The mic input will be around 2K ohm. The bass roll off into this load indicates too small a preamp output coupling cap for this impedance. If you use the preamp with your other amps, which might have have lowish impedance inputs, then I am not surprised they sound anemic in the bass.
If the other amplifiers include preamps and have lower output impedance, the downward tilt and bumps will disappear. The sound will be leaner. I have a valve amp and its output impedance reacts with the speakers impedance to give a punchier sound in the bass. It is not compression.
The output impedance of the preamp at LF seems high. The mic input will be around 2K ohm. The bass roll off into this load indicates too small a preamp output coupling cap for this impedance. If you use the preamp with your other amps, which might have have lowish impedance inputs, then I am not surprised they sound anemic in the bass.
If the other amplifiers include preamps and have lower output impedance, the downward tilt and bumps will disappear. The sound will be leaner. I have a valve amp and its output impedance reacts with the speakers impedance to give a punchier sound in the bass. It is not compression.
Last edited:
The “bumpiness” of the response looks like what you get from tone control inaccuracy. Possibly the use of audio taper pots in a Baxandall circuit - almost impossible to maintain a flat response in the “zero” position when you do that. It seems to be happening in the power amp circuit - I’ve seen the tone controls used on the power amp feedback before to save one miserable preamp stage that couldn’t add more than 50 cents to the BOM. Cheapskates. More inaccuracy with the speaker connected means the open loop gain drops under load, which is typical.
The slope in the preamp is probably just to make it sound bassy. The corner at the low end is just due to coupling caps, intentionally set to roll off response below 20.
The slope in the preamp is probably just to make it sound bassy. The corner at the low end is just due to coupling caps, intentionally set to roll off response below 20.
Classic circuit with DC output. Baxandall timbre. Loudness. Universal preamp with nfb switching.
Last edited:
A lot bumpier than THAT if the loudness contour were “ON”. Often 20 dB or more bass boost close to minimum volume setting. A common trick on those old receivers was to run the preamp signal HOT so that the amp put out FULL power in the 11:00 position of the volume control. FM was so compressed that this almost always happened on the rock/pop stations at least. So the loudness was ALWAYS boosting the bass a lot with any reasonable volume setting.
This receiver is typical of the 1970's/80's style of design.
And Nikko, Harman Kardon, etc, all had the same seductive type of response, pleasing to the ears (of the general public).
My HK 330A is like that, and having only reasonable wattage ratings, it bellows out that rich-type (impressive) sound.
And even today, these things are coveted and in demand by people.
There is something to be said for the non-linear, less than purist, type of sound.
And Nikko, Harman Kardon, etc, all had the same seductive type of response, pleasing to the ears (of the general public).
My HK 330A is like that, and having only reasonable wattage ratings, it bellows out that rich-type (impressive) sound.
And even today, these things are coveted and in demand by people.
There is something to be said for the non-linear, less than purist, type of sound.
Thanks for the replies, gentlemen. Very insightful replies.
Spladski, thank you for your replies. The 4dB was a terrible typo on my part.
Your comment about the coupling cap being too small for the mic input impedance got me looking at a paper copy of the schematic. On the output of the preamp, there is 0.47uF and 1k. On the power amp input there is 100k to the ground plane, then a 0.47uF cap and 2.2k to the first 2SA539, and between the 2.2k and 2SA539 is 3.3k to the ground plane. High output Z and weird input Z.
Spladski, thank you for your replies. The 4dB was a terrible typo on my part.
Your comment about the coupling cap being too small for the mic input impedance got me looking at a paper copy of the schematic. On the output of the preamp, there is 0.47uF and 1k. On the power amp input there is 100k to the ground plane, then a 0.47uF cap and 2.2k to the first 2SA539, and between the 2.2k and 2SA539 is 3.3k to the ground plane. High output Z and weird input Z.
Nikko made it that way for a reason....
There is nothing worthy of musical enjoyment except for power-robbing subsonics if an amp is built with large capaciters in the signal path.
Tailoring impedences and capacitors helps to gain more power where the music really is.