Hi Guys,
I have a pair of these both operating in mono bridged, on for left and one for right.
I've had the pair around 6 months and lately I've noticed a difference in volume between the 2 when the attenuators on both amps are set to the middle indent.
I can't say if this is a difference which has developed over time or I've only just started to notice it.
So I opened them both up and hit them with a 6khz sine wave at the same input level, to the left input for mono operation. Sure enough one amp is getting 24.7 volts at the speaker terminals when the attenuator is at the middle indent and the other reads 30 volts. I tried a few seeps of the pots up and down, getting 100 volts for both on maximum. (I also tested in stereo mode and sure enough the voltage exactly halves as expected on both amps)
First thing I did was perform the idle bias calibration, they were both a tiny bit out (give or take 0.1 - 0.3 mv) so I adjusted them and get a steady 7mv after about 30 mins idling.
I then retested the output voltages but I'm still getting the same ~5 volts difference between the 2 amps when set to the mid position on the attenuator.
Aside from the volume differences both amps sound equally good, I have pretty good ears and I can't hear anything to make me think there are any issues with either amp in that regard. They are also both near silent with a barely detectable amount of hiss when at full volume with nothing being played, also neither have any hum what soever coming from the trnsformer.
So I guess my first 3 questions are
1: what voltage should i be getting at the output (i couldn't find anything on line specifically regarding this)
2: could the difference simply be down to the transistors and the amps them selves were never intended to put out an exactly matched voltage?
3: has anyone encountered anything like this before who can offer any advice on where I should start my trouble shooting?
I have a pair of these both operating in mono bridged, on for left and one for right.
I've had the pair around 6 months and lately I've noticed a difference in volume between the 2 when the attenuators on both amps are set to the middle indent.
I can't say if this is a difference which has developed over time or I've only just started to notice it.
So I opened them both up and hit them with a 6khz sine wave at the same input level, to the left input for mono operation. Sure enough one amp is getting 24.7 volts at the speaker terminals when the attenuator is at the middle indent and the other reads 30 volts. I tried a few seeps of the pots up and down, getting 100 volts for both on maximum. (I also tested in stereo mode and sure enough the voltage exactly halves as expected on both amps)
First thing I did was perform the idle bias calibration, they were both a tiny bit out (give or take 0.1 - 0.3 mv) so I adjusted them and get a steady 7mv after about 30 mins idling.
I then retested the output voltages but I'm still getting the same ~5 volts difference between the 2 amps when set to the mid position on the attenuator.
Aside from the volume differences both amps sound equally good, I have pretty good ears and I can't hear anything to make me think there are any issues with either amp in that regard. They are also both near silent with a barely detectable amount of hiss when at full volume with nothing being played, also neither have any hum what soever coming from the trnsformer.
So I guess my first 3 questions are
1: what voltage should i be getting at the output (i couldn't find anything on line specifically regarding this)
2: could the difference simply be down to the transistors and the amps them selves were never intended to put out an exactly matched voltage?
3: has anyone encountered anything like this before who can offer any advice on where I should start my trouble shooting?
Nice amps 
In 2 independent stereo amplifiers, it is unlikely that you are going to get matched output levels for the same positional setting of the volume controls anyway. I noticed this in a showroom many years ago, where 2 stereo amps with close serial numbers were set up for closely balanced output with top-notch instruments. The volume control settings were also noticeably different which started my thinking about the reality of our hearing. It happens that aural sensitivity is logarithmic and that is how significant tracking errors in the potentiometers are often masked. The 21% electrical difference in your case, is only minor in aural terms.
Having both volume controls ganged precisely together requires a close and expensive degree of matching of the wafers and rotor to within the tolerance range you expect. Volume controls should ideally also follow a logarithmic law and this where it gets awkward, since it's not possible to make an affordable carbon log. track with any accuracy and the variation between parts can be significant in resistance value, hence also the signal levels.
6kHz may also be too high a frequency for any multimeter short of a true RMS type or a designated audio millivolt meter. I find that budget meters roll off their response from below even 4kHz. Whilst this won't affect direct comparisons using the same meter, it means your measurements won't be relevant to specifications or other external data. Try a frequency between say, 400Hz and 1kHz to be more certain.
The amplifiers may not require trouble shooting at all but if there really has been a fall-off in response over the six months you've owned them, it won't be in the semis. You would be looking for trouble in parts that deteriorate which boils down to the electrolytic capacitors in the low level signal path. These often begin to fail after about 20 years and that is about the age of the these models, made from 1989-1999. Sony TA-N55ES - Manual - Stereo Power Amplifier - HiFi Engine
In 2 independent stereo amplifiers, it is unlikely that you are going to get matched output levels for the same positional setting of the volume controls anyway. I noticed this in a showroom many years ago, where 2 stereo amps with close serial numbers were set up for closely balanced output with top-notch instruments. The volume control settings were also noticeably different which started my thinking about the reality of our hearing. It happens that aural sensitivity is logarithmic and that is how significant tracking errors in the potentiometers are often masked. The 21% electrical difference in your case, is only minor in aural terms.
Having both volume controls ganged precisely together requires a close and expensive degree of matching of the wafers and rotor to within the tolerance range you expect. Volume controls should ideally also follow a logarithmic law and this where it gets awkward, since it's not possible to make an affordable carbon log. track with any accuracy and the variation between parts can be significant in resistance value, hence also the signal levels.
6kHz may also be too high a frequency for any multimeter short of a true RMS type or a designated audio millivolt meter. I find that budget meters roll off their response from below even 4kHz. Whilst this won't affect direct comparisons using the same meter, it means your measurements won't be relevant to specifications or other external data. Try a frequency between say, 400Hz and 1kHz to be more certain.
The amplifiers may not require trouble shooting at all but if there really has been a fall-off in response over the six months you've owned them, it won't be in the semis. You would be looking for trouble in parts that deteriorate which boils down to the electrolytic capacitors in the low level signal path. These often begin to fail after about 20 years and that is about the age of the these models, made from 1989-1999. Sony TA-N55ES - Manual - Stereo Power Amplifier - HiFi Engine
Hey Ian, thanks for the sensible response.
You've got me thinking about doing some frequency response test on them just to be certain that they are similar. I do music production so I have the know how and facilities to sweep them and capture the output, my only question is what's the best way to reduce the voltage of the output from the speaker terminals to line level so i can feed them into an audio interface? Is it as simple as putting some appropriate resistors on the outputs that can handle those voltages? Bear in mind it's only going to be for a brief period of time and not at a high output volume.
Does this sound sensible or is there a better way to do it?
You've got me thinking about doing some frequency response test on them just to be certain that they are similar. I do music production so I have the know how and facilities to sweep them and capture the output, my only question is what's the best way to reduce the voltage of the output from the speaker terminals to line level so i can feed them into an audio interface? Is it as simple as putting some appropriate resistors on the outputs that can handle those voltages? Bear in mind it's only going to be for a brief period of time and not at a high output volume.
Does this sound sensible or is there a better way to do it?
I agree with Ian.
measure the level after the gain/volume pots. or measure the resistance of the pots on the same setting. I think that there lies the difference.
Most amplifiers with seperate gain pot per channel require the gain set to maximum when bridged to minimize difference between the resistance tracks of one single potentiometer.
+/-20% resistance spec between pots is not uncommon. Even with the big ALPS types
RK27112A00AK | Alps Alpine RK27112A00AK 2 Gang Rotary Carbon Potentiometer with a 6 mm Dia. Shaft, 100kΩ, +-20%, 0.05W, Logarithmic | RS Components
PTV142B20E120AA104 | Bourns PTV142B20E120AA104 2 Gang Rotary Carbon Potentiometer with a 6 mm Dia. Shaft, 100kΩ, +-20%, 0.05W, Audio | RS Components
I'd not start blindly swapping components when they gave you a volume control to make adjustments between amplifiers. More so because, if i understand correctly, measurements between channels on one amplifier are equal and bridging works as intended.
measure the level after the gain/volume pots. or measure the resistance of the pots on the same setting. I think that there lies the difference.
Most amplifiers with seperate gain pot per channel require the gain set to maximum when bridged to minimize difference between the resistance tracks of one single potentiometer.
+/-20% resistance spec between pots is not uncommon. Even with the big ALPS types
RK27112A00AK | Alps Alpine RK27112A00AK 2 Gang Rotary Carbon Potentiometer with a 6 mm Dia. Shaft, 100kΩ, +-20%, 0.05W, Logarithmic | RS Components
PTV142B20E120AA104 | Bourns PTV142B20E120AA104 2 Gang Rotary Carbon Potentiometer with a 6 mm Dia. Shaft, 100kΩ, +-20%, 0.05W, Audio | RS Components
I'd not start blindly swapping components when they gave you a volume control to make adjustments between amplifiers. More so because, if i understand correctly, measurements between channels on one amplifier are equal and bridging works as intended.
Thanks,
yes, what you are both saying makes perfect sense, given that both of the amps measured exactly 100V (well one was 100 the other 99.8) with the pot at max
I'll measure the pots this weekend for peace of mind.
thanks for your input.
For some background on typical audio pots and what you could expect to find, look at taper "A" for example in Alpha's datasheet.
Potentiometer's Tapers
It illustrates what a typical "log taper" volume pot. is actually like, since the taper is only approximated by 2 or more linear sections spliced together when deposited on the wafer surface. The rotor contact follows a circular path across those linear sections in a crude, inconsistent approximation to a logarithmic law, but with break point(s) right where your listening levels are likely to be. Even so, this has worked acceptably for general audio purposes for decades. Just don't expect any reference scale around the knobs to be more than decoration
Potentiometer's Tapers
It illustrates what a typical "log taper" volume pot. is actually like, since the taper is only approximated by 2 or more linear sections spliced together when deposited on the wafer surface. The rotor contact follows a circular path across those linear sections in a crude, inconsistent approximation to a logarithmic law, but with break point(s) right where your listening levels are likely to be. Even so, this has worked acceptably for general audio purposes for decades. Just don't expect any reference scale around the knobs to be more than decoration
So last night it came to me why I've only recently noticed this in balance between the 2 amp.
I used to have the amps set very low with just my dac going directly into the amps, so I'd be manually judging where to set the pots so they were balanced. Last month I added a passive 2 in switch box with an attenuator. So I just set the amps to the middle indent assuming that was accurate and just use the passive pre to knock it down to an acceptable level (theses amps are way too loud even at the half way point)
So obviously the tolerance difference between the pots on each amp is obviously more apparent at this mid point than it was much lower down at say 10 percent.
Mystery solved.
So I've just hit them both with a sine wave and manually balanced the attenuated positions so they are outputting the same voltage.
Thanks for the input
I used to have the amps set very low with just my dac going directly into the amps, so I'd be manually judging where to set the pots so they were balanced. Last month I added a passive 2 in switch box with an attenuator. So I just set the amps to the middle indent assuming that was accurate and just use the passive pre to knock it down to an acceptable level (theses amps are way too loud even at the half way point)
So obviously the tolerance difference between the pots on each amp is obviously more apparent at this mid point than it was much lower down at say 10 percent.
Mystery solved.
So I've just hit them both with a sine wave and manually balanced the attenuated positions so they are outputting the same voltage.
Thanks for the input
When I was doing the individual measurement of the amps that did not have the passive pre in the chain so I was taking the true measurement from each amp.
I will however calibrate the whole system with the passive pre in circuit to ensure I'm getting equal out puts form both amps with everything in circuit.
I will however calibrate the whole system with the passive pre in circuit to ensure I'm getting equal out puts form both amps with everything in circuit.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.