I would expect a lot better performance for a $25k line-level box of gain, but I'll admit to being unreasonable.
Hi,
Just about everything I get into is iceing on the cake as most amps are very good by now. It's the 1% left to do. However, the curves I showed are the type of amps which have high distortion above audio. Can you find power amp with similar curves - beg, borrow or steal. These are Power Amps using high GNFB. Rising distortion as GNFB is falling. Find those to do the tests.
Cross-over notch isnt the same affect as from HF/RF as the continuously rising THD of a power amplifier above 20KHz. At the rate many of these amplifers THD rises, ill be putting out some gross THD numbers above 20KHz.
THx-RNMarsh
Last edited:
OPPO notes --- the sound was lacking in clarity in the lower bass. I play bass and the voltage is relatively higher across caps there. My load Z is 9K.
It reminded me of high DA caps, like electrolytic. I asked JC what kind of caps does it have if any coupling caps. He told me the brand and also sent me the schematic.
The analog input and output of opamps are cap coupled with polar (yes, polar) electrolytics. For such a product, couldnt they pay for Bi-Polar types? Where have they been. Maybe the digital or video guys designed the analog as well. I opened it up and replaced them with bipolar and sure enough the bass was clean and more real (accurate). A slight haze over everything was less.
The upper mid to treble is still CD sounding..... less and less clear as the freq goes up. The fundemental freqs are all there but not the quality. A cymbol has a sheen and sparkle that is very clear. I can tell its a cymbal but it does not sound iike a real cymbal... that sparkle and shimmer and bright clear sound isnt there.... smeared and less distinct. [missing harmonics, G-D, BW, filters, etc. ???]
Done. Next up USB 24/96 HD downloads thru OPPO.
THx-RNMarsh
It reminded me of high DA caps, like electrolytic. I asked JC what kind of caps does it have if any coupling caps. He told me the brand and also sent me the schematic.
The analog input and output of opamps are cap coupled with polar (yes, polar) electrolytics. For such a product, couldnt they pay for Bi-Polar types? Where have they been. Maybe the digital or video guys designed the analog as well. I opened it up and replaced them with bipolar and sure enough the bass was clean and more real (accurate). A slight haze over everything was less.
The upper mid to treble is still CD sounding..... less and less clear as the freq goes up. The fundemental freqs are all there but not the quality. A cymbol has a sheen and sparkle that is very clear. I can tell its a cymbal but it does not sound iike a real cymbal... that sparkle and shimmer and bright clear sound isnt there.... smeared and less distinct. [missing harmonics, G-D, BW, filters, etc. ???]
Done. Next up USB 24/96 HD downloads thru OPPO.
THx-RNMarsh
Last edited:
9k should not make any problem, if the output had at least moderate quality. These consumer products are boring ...
Richard, you can probably bypass the output cap. The offset might still be very low. Your load Z is lower than normal consumer stuff. 50K is more usual.
Richard, do you mean the power amp input? Stereophile found the input Z for the Benchmark AHB2 to be much higher, but maybe you are talking about something else and I missed it earlier.
For the last power amp I was involved with, a switchmode amp which began as a IR reference design, I used a JFET input follower with load compensation and a d.c. servo, and had a d.c. input termination resistor of 1M. This would have been o.k. except I wasn't present when two other people in the now-defunct company decided to test it. They used clip leads to hook up to the input (!). Input-output coupling induced oscillations, and they managed to turn it off before much damage was done, but the output Zobels were smoking. This led to yet more parts to detect oscillations....
The 9K loading will not necessarily affect the output IC, but it might load the cap. We are talking about SUBJECTIVE here, not a meter reading.
Richard, I am now told that there is a DELIBERATE OFFSET added to the main output stage. You might have to find and remove it as well.
Now That is interesting. Why? does it help the polar cap? and why not just use a bipolar? Or helps the opamp in some way?
The output of the OPPO goes to the input on the BenchMak-2.
I am not interested enough in CD to do any other mods to it. It is an above average performer. But I do like to use HD mastered 24/96 music files on USB memory stick and listen. There is a USB port on front and rear panel for this 😎
THx-RNMarsh
Last edited:
I use an OPPO to play CDs and USB hidef downloads, but I use the Oppo's digital output to feed an external DAC - the external DAC uses Borbely's discrete jfet IV stage direct coupled/servo offset controlled powered by shunt regulators, the DAC chips are powered by Jung regulators.
There is a large improvement in resolution, dynamics, bass control, over the internal op amp / 3 terminal regulator powered analog stages found in most players, and while the OPPO analog stage is better than most it too is bested by superior discrete IV/ gain stages and better power supplies. It is evident on any CD that's played back, not just on the best high res material.
There is a large improvement in resolution, dynamics, bass control, over the internal op amp / 3 terminal regulator powered analog stages found in most players, and while the OPPO analog stage is better than most it too is bested by superior discrete IV/ gain stages and better power supplies. It is evident on any CD that's played back, not just on the best high res material.
Richard,
In this schematic Audiophile Musings: Another OPPO Mod you can see where the offset voltage is connected to (look at U24 and 25). I don't know what level the offset voltage is at, but I presume it is high enough to keep the polars c93 and 94 happy. If this is the case, all the voltage swing may be kept in the positive half of the opamp's potential. It's a bit of a guess, you would have to measure. Question for me is: would such a setup have similar effect as biasing the opamp into class A by pulling a constant current from the output, that is, to prevent crossover switching between the output tors? That would seem to be the case.
In this schematic Audiophile Musings: Another OPPO Mod you can see where the offset voltage is connected to (look at U24 and 25). I don't know what level the offset voltage is at, but I presume it is high enough to keep the polars c93 and 94 happy. If this is the case, all the voltage swing may be kept in the positive half of the opamp's potential. It's a bit of a guess, you would have to measure. Question for me is: would such a setup have similar effect as biasing the opamp into class A by pulling a constant current from the output, that is, to prevent crossover switching between the output tors? That would seem to be the case.
Last edited:
- Status
- Not open for further replies.