Considering a soon to be released major name brand DAC w/XLR and RCA outputs, as follows:
RCA > Dayton Audio kilowatt sub amp, unbalanced 12k Ohm input, 750mV for full output
XLR > Hypex NCore NC400BTL, fully balanced 50k Ohm input, 2.0V for full output
In reply to my concern about the above load, the DAC designer replied: "One OPAMP buffer drives both the RCA and XLR outputs. The RCA signal = XLR pin 2. A 75 Ohm R drives the RCA. A 110 Ohm R drives the XLR. The OPAMP delivers >25mA, preventing interaction/interference between the XLR and RCA loads."
Does above source appear to pass adequate current for the bass amp load? What is potential for above source to lack current for the bass amp, thus softening bass dynamics?
XLR pin2 load is about 10k Ohm (sub amp) while XLR pin3 load is 50k Ohm. Does this disparity not audibly degrade Hypex amp performance?
RCA > Dayton Audio kilowatt sub amp, unbalanced 12k Ohm input, 750mV for full output
XLR > Hypex NCore NC400BTL, fully balanced 50k Ohm input, 2.0V for full output
In reply to my concern about the above load, the DAC designer replied: "One OPAMP buffer drives both the RCA and XLR outputs. The RCA signal = XLR pin 2. A 75 Ohm R drives the RCA. A 110 Ohm R drives the XLR. The OPAMP delivers >25mA, preventing interaction/interference between the XLR and RCA loads."
Does above source appear to pass adequate current for the bass amp load? What is potential for above source to lack current for the bass amp, thus softening bass dynamics?
XLR pin2 load is about 10k Ohm (sub amp) while XLR pin3 load is 50k Ohm. Does this disparity not audibly degrade Hypex amp performance?
If you go purely by "specs", not a problem. But, in practice- PROBLEM. Just went through this with a client recently, after discovering, to our surprise, that his Revel sub amps have an absurdly low 14k input. Apparently, it's become common, but inexcusable, practice to have low input resistance specs on sub amps, which is APPALLINGLY STUPID, and there is NO rational reason to make it so low. Technically, the average opamp or otherwise pre/dac output stage can handle <20k load without overt measured distortion, but, sonically, going that low takes the "guts" out of the bass dynamics & often overall tone. Since there is no justifiable reason to have a low input resistance, any decent designer will give a sub amp at LEAST 100k input, and 200k if they're really wise. This is because, obviously, the sub amp input is always connected in parallel with the main amp, which will usually have a 50-100k input. With sub amp load high, it'll keep the total load on the preamp well over 25k, comfortable for all but the saddest output stages.
So, if I were you, I'd either open up the sub amp & change it's input Z to at least 100k, or, if you're not very skilled, at least put a >47k resistor in series with the cable feed to the sub, externally. You will most likely have enough gain range on the sub amp to make up for the attenuation thus introduced.
Have to say that the response of the dac designer shows a surprising lack of both practical & technical knowledge about the subject.
So, if I were you, I'd either open up the sub amp & change it's input Z to at least 100k, or, if you're not very skilled, at least put a >47k resistor in series with the cable feed to the sub, externally. You will most likely have enough gain range on the sub amp to make up for the attenuation thus introduced.
Have to say that the response of the dac designer shows a surprising lack of both practical & technical knowledge about the subject.
Stephen,
Very happy to read your reply, which is consistent with my thinking.
Do two different loads between pins 2/3 degrade main amp performance?
I presume it's best to locate the 47k R as close as possible to the sub amp input? What is estimated input sensitivity w/this R (OEM is 750mV).
Do certain internal sub amp mods improve performance beyond the R?
Is there no design benefit for any amp input impedance to be <100k Ohm?
I performed AB test with a particular unbalanced preamp output: first driving 47k Ohm power amp input, then two of the same inputs in parallel for 23.5k Ohm. The latter has less dynamic impact and less bass power. That's why I asked in the first place.
I have no idea why sub amps seem to have much lower input Z than the average amp, but it definitely appears to be the case. And I echo your sentiment: I suspect persons may audition such rigs w/source incapable of properly driving the resulting load in parallel with the main amp. I think this is one of those elephants in the audiophile closet. Like an OTL amp driving old Apogee ribbons, but less severe than that...
Old Naim and Spectral amps had unusually low input Z, 10k Ohm IIRC.
Very happy to read your reply, which is consistent with my thinking.
Do two different loads between pins 2/3 degrade main amp performance?
I presume it's best to locate the 47k R as close as possible to the sub amp input? What is estimated input sensitivity w/this R (OEM is 750mV).
Do certain internal sub amp mods improve performance beyond the R?
Is there no design benefit for any amp input impedance to be <100k Ohm?
I performed AB test with a particular unbalanced preamp output: first driving 47k Ohm power amp input, then two of the same inputs in parallel for 23.5k Ohm. The latter has less dynamic impact and less bass power. That's why I asked in the first place.
I have no idea why sub amps seem to have much lower input Z than the average amp, but it definitely appears to be the case. And I echo your sentiment: I suspect persons may audition such rigs w/source incapable of properly driving the resulting load in parallel with the main amp. I think this is one of those elephants in the audiophile closet. Like an OTL amp driving old Apogee ribbons, but less severe than that...
Old Naim and Spectral amps had unusually low input Z, 10k Ohm IIRC.
Last edited:
12k in parallel with 50k is 9.7k. Almost any opamp can drive a few volts into 9.7k without any difficulty.
If the opamp can't deliver the current, the effect will not be on "bass dynamcs" but overall distortion, or possibly treble distortion.
Do some arithmetic. 2V RMS (the typical DAC output) is 2.8V pk. 2.8V across 9.7k is 0.288mA. The designer claims that his output stage can deliver 25mA. That is 38dB of headroom.
If you know enough to mod the bass amp, then you know enough to realise that it doesn't need modding.
If the opamp can't deliver the current, the effect will not be on "bass dynamcs" but overall distortion, or possibly treble distortion.
Do some arithmetic. 2V RMS (the typical DAC output) is 2.8V pk. 2.8V across 9.7k is 0.288mA. The designer claims that his output stage can deliver 25mA. That is 38dB of headroom.
If you know enough to mod the bass amp, then you know enough to realise that it doesn't need modding.
OK. Thanks very much for the math, which I greatly appreciate. I prefer to avoid needless work in this matter.
I respectfully request your opinion on a real life AB test w/two 2003-04 Pioneer VSX-D912 HT receivers, pre-HDMI era. The "912" pure analog preamp stage sounded indistinguishable from my $7500 SST Ambrosia line stage (2012 TAS Product of the Year). The 912 power amp stage is a low-current pure analog chip. Powering ideal loads (>12 Ohm, easy phase angle), the 912s power amp stage is exemplary, thoroughly more refined than a Plinius SA-50 Mk3 pure class A power amp.
The entire test was single channel, not stereo.
Test A:
CDP unbalanced analog output (outperformed Parasound Halo CD1) > #1 912 pure analog input "DVD multi-ch 7.1," preamp output > #2 912 pure analog input "DVD multi-ch 7.1," speaker output > highly refined full range speaker
In Test A, #1 912 preamp output drove one input on #2 912. Test B is identical except #1 912 preamp output drove two inputs on #2 912.
Test B performance audibly degraded.
Pioneer omits specs for "DVD Multi-Ch 7.1" pure analog pass through and preamp outputs employed in the above described test. The closest listed specs are below, but listed "Inputs" pass through the ADC/DAC, and listed "Outputs" are fixed level tape "REC," none of which I ever used.
"...Input Sensitivity/Impedance CD, VCR/DVR, CD-R/TAPE/MD, DVD/LD, TV/SAT: 200 mV/47 kΩ
Output Level/Impedance VCR/DVR REC, CD-R/TAPE/ MD REC: 200 mV/2.2 kΩ..."
Do the test results surprise readers? Potential explanations appreciated.
I respectfully request your opinion on a real life AB test w/two 2003-04 Pioneer VSX-D912 HT receivers, pre-HDMI era. The "912" pure analog preamp stage sounded indistinguishable from my $7500 SST Ambrosia line stage (2012 TAS Product of the Year). The 912 power amp stage is a low-current pure analog chip. Powering ideal loads (>12 Ohm, easy phase angle), the 912s power amp stage is exemplary, thoroughly more refined than a Plinius SA-50 Mk3 pure class A power amp.
The entire test was single channel, not stereo.
Test A:
CDP unbalanced analog output (outperformed Parasound Halo CD1) > #1 912 pure analog input "DVD multi-ch 7.1," preamp output > #2 912 pure analog input "DVD multi-ch 7.1," speaker output > highly refined full range speaker
In Test A, #1 912 preamp output drove one input on #2 912. Test B is identical except #1 912 preamp output drove two inputs on #2 912.
Test B performance audibly degraded.
Pioneer omits specs for "DVD Multi-Ch 7.1" pure analog pass through and preamp outputs employed in the above described test. The closest listed specs are below, but listed "Inputs" pass through the ADC/DAC, and listed "Outputs" are fixed level tape "REC," none of which I ever used.
"...Input Sensitivity/Impedance CD, VCR/DVR, CD-R/TAPE/MD, DVD/LD, TV/SAT: 200 mV/47 kΩ
Output Level/Impedance VCR/DVR REC, CD-R/TAPE/ MD REC: 200 mV/2.2 kΩ..."
Do the test results surprise readers? Potential explanations appreciated.
Last edited:
DF96 is still stuck in the old "if it measures good it must sound good" school.
In practice. the transient dynamics and other nuances of real sound suffer WAY sooner than measured specs suggest in the area we're discussing. Sad fact is the VERY few, as in barely any, preamps or dacs or other line level output stages are remotely close to adequately power-supplied(meaning stored energy right at the stage, not just in supply area) to have any chance of driving a line, let alone a low-Z load at the end of that line, well enough to preserve all nuances of music. Even "pro audio" gear that has outputs designed to drive 600ohms pretty much never are designed well enough to really do the job anywhere near perfectly. So, unless you're like me, in the habit of designing discrete output stages with totally overkill power supply support, you just have to accept that you need to keep every line load, ideally, well over 47kohm to have a chance at great sound.
And it was well-thought, on your part, to suggest having the series resistor for the sub amp feed at the amp end of the cable line. This assures that at least the line has the max available current drive from the source.
In practice. the transient dynamics and other nuances of real sound suffer WAY sooner than measured specs suggest in the area we're discussing. Sad fact is the VERY few, as in barely any, preamps or dacs or other line level output stages are remotely close to adequately power-supplied(meaning stored energy right at the stage, not just in supply area) to have any chance of driving a line, let alone a low-Z load at the end of that line, well enough to preserve all nuances of music. Even "pro audio" gear that has outputs designed to drive 600ohms pretty much never are designed well enough to really do the job anywhere near perfectly. So, unless you're like me, in the habit of designing discrete output stages with totally overkill power supply support, you just have to accept that you need to keep every line load, ideally, well over 47kohm to have a chance at great sound.
And it was well-thought, on your part, to suggest having the series resistor for the sub amp feed at the amp end of the cable line. This assures that at least the line has the max available current drive from the source.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.