Personally, I've seen using a variety of OPAMP, OP275 price when you consider the part that I like it.
OPA637 and OPA627 or applying expensive chips for audio equipment and try both, but both look good as advanced equipment and cheap OP285 and OP275, or would think we could make.
Aspects when considering the specification LME49720 I'm getting good chips.
OPA637 and OPA627 or applying expensive chips for audio equipment and try both, but both look good as advanced equipment and cheap OP285 and OP275, or would think we could make.
Aspects when considering the specification LME49720 I'm getting good chips.
As for music and the "not sounding thin" and considering that your average opamp can't get the job done, then you're probably not after the lowest distortion figures. A vacuum tube may be a perfect candidate. However, there are some rare solid state options. Moosefet, LM1875, and the discrete preamp at Decibel Dungeon provide some examples. The gain ranges are all different though.
Its also possible that you just need some higher current capacity along with your voltage amplification. The Creative Labs X-FI cards fall flat in comparison to the Via based sound cards just for this very reason. A preamp could possibly solve this, but. . .
Perhaps whatever solution works for you could also work for my problem of "my preamp always seems to need. . . a preamp."
Good news! Put in the 2nd amp (modded 3 for my car setup) with dual OPA627 which had been running 24/7 for the last 3 weeks in place of the NJM353D to drive the highs and midrange, shifted the previous to drive the midbass and my entire setup sounded exactly what many said of it .... extremely clean sounding, detailed and very transparent. Top-hat, cymbal etc. sounds so clear and transparent ... to the effect of being slightly bright ... details are as usual ... not dark or slow now but tad less musical than before which is surprising. This is probably also "helped" by population of 8 pcs of 10uF 50V Blackgate N capacitors in place of Nichicon NP BP as input coupling capacitors and 4 pcs of the same to replaced Nichicon Polar FA which should be for some bypass function. Comparatively, the original amp was veiled although nicely veiled sounding but musical due and it should even be better after some EQ since the previous tuning is for the original amps. However, I can also understand Andrea's comment on OPA627 being sterile (overly clinical due to it clean sounding nature) since I believe it can be if not match with warm sound equipment as in my case of Class A, Zero Feedback topology amp. I think I will need to run-in the 1st modded amp for another 1 weeks to achieve about 500 hours and the initial poor SQ may be due to the notoriously known "long burn-in" nature of the Blackgate N caps !! A-B is done with another 3 Phonon amp refurnished in all original Nichicon capacitors which had been the resident setup.
May be off topic but for understanding the slight brightness, it may be due to characteristic of main Alpine equipment:
Alpine CDA-7998R, PXA-H900, DVA-S680E, CHA-S609, TUA-T100DAB, Phonon 340C (3X), Micro Precision z-series Tweeters, Z-series Mid and SEAS reference midbass, Rainbow Vanadium 8" Sub, Belden speaker cable, Mogami IC.
Not more fiddling with op amp for now and it is tough with the PCB mounted from bottom up with the power mosfets and transistors in contact with the main single pc aluminum block chassis cum heatsink !! In conclusion, OPA627 is find for my 15 year old amps and it is a big relief ....
For the original amp sound with NJM353, I was very happy with the Nichicon capacitors especially with real sounding and analog-ish vocals as well as musical sounding thru the exceptional micro-precision Z-series midrange driver. However, I always felt that clarity and transparency is somewhat lacking compared to some current top-of-range amps and setup ... hence was moved to try the OPA627 due to many reviews concluding that it is clean sounding but yet transparent review ... should be better than the "jellybean class" of dated 90s' NJM353.
Regards,
Richard
May be off topic but for understanding the slight brightness, it may be due to characteristic of main Alpine equipment:
Alpine CDA-7998R, PXA-H900, DVA-S680E, CHA-S609, TUA-T100DAB, Phonon 340C (3X), Micro Precision z-series Tweeters, Z-series Mid and SEAS reference midbass, Rainbow Vanadium 8" Sub, Belden speaker cable, Mogami IC.
Not more fiddling with op amp for now and it is tough with the PCB mounted from bottom up with the power mosfets and transistors in contact with the main single pc aluminum block chassis cum heatsink !! In conclusion, OPA627 is find for my 15 year old amps and it is a big relief ....
For the original amp sound with NJM353, I was very happy with the Nichicon capacitors especially with real sounding and analog-ish vocals as well as musical sounding thru the exceptional micro-precision Z-series midrange driver. However, I always felt that clarity and transparency is somewhat lacking compared to some current top-of-range amps and setup ... hence was moved to try the OPA627 due to many reviews concluding that it is clean sounding but yet transparent review ... should be better than the "jellybean class" of dated 90s' NJM353.
Regards,
Richard
From audio-cube (NL) site:
"The SILMIC II is best for reproducing different timbres, natural colors and warmth in a recording. It offers most presence in the low-mid range with lots of texture (tangible). Bass is round and warm, the mid and highrange are very refined, 'silky', quiet, but with lots of openess (more open than CERAFINE, less open than Black Gate FK).
The CERAFINE is also very good for reproducing different timbres, natural colors, but has a more direct sound than SILMIC (less direct than Black Gate FK). Bass is round and warm, the midrange has most presence and the high range is lively.
The Black Gate FK has the most open and clean sound, with the main presence in the high range. It sounds thinner, offers somewhat less texture and timbre. It offers most pulse, attack in both the low, mid and high ranges."
To me your description fits perfectly on this description of BGs.
Consider also that both OPA627 and OPA2132 do change their timbre according to the value of the PS bypass caps.
The value I use mostly is 47uF but sometimes it should be tuned to the circuit.
If you used 10uF also for PS bypass, 10uf is too low.
I do agree that the "slight brightness" is probably due to other gear that preceed the amp, I've had the same experience with my SACD player.
If I were in your shoes I would replace the Blackgates with Silmics, particularly the PS ones which requires, IMHO, a higher value (47uf should be good).
If you choose to replace caps consider that also Silmics needs at leat 50 hours to stabilize their timbre, at first bass is too much.
THD is not an interesting figure in audio. It doesn't tell how an amp sounds. I've experimented with discrete curcuits in different ways. If open loop gain is lowered, the THD will rise, but the amp doesn't sound worse. I think it sounds more or less the same. Tweaks that lower THD don't automatically yield a better sound. Tweaks that leave THD unchanged can have "major" effects.
From audio-cube (NL) site:
"The SILMIC II is best for reproducing different timbres, natural colors and warmth in a recording. It offers most presence in the low-mid range with lots of texture (tangible). Bass is round and warm, the mid and high range are very refined, 'silky', quiet, but with lots of openess (more open than CERAFINE, less open than Black Gate FK).
The CERAFINE is also very good for reproducing different timbres, natural colors, but has a more direct sound than SILMIC (less direct than Black Gate FK). Bass is round and warm, the midrange has most presence and the high range is lively.
The Black Gate FK has the most open and clean sound, with the main presence in the high range. It sounds thinner, offers somewhat less texture and timbre. It offers most pulse, attack in both the low, mid and high ranges."
To me your description fits perfectly on this description of BGs.
Consider also that both OPA627 and OPA2132 do change their timbre according to the value of the PS bypass caps.
The value I use mostly is 47uF but sometimes it should be tuned to the circuit.
If you used 10uF also for PS bypass, 10uf is too low.
I do agree that the "slight brightness" is probably due to other gear that preceed the amp, I've had the same experience with my SACD player.
If I were in your shoes I would replace the Blackgates with Silmics, particularly the PS ones which requires, IMHO, a higher value (47uf should be good).
If you choose to replace caps consider that also Silmics needs at leat 50 hours to stabilize their timbre, at first bass is too much.
I love your description of voicing at the power circuit. That is a fun sport indeed.
The description of Silmic is either a poor capacitor with an hf failure mode or it is an overlarge capacitor; however, both can be well employed in power circuits.
The description of Cerafine is simply the effect of too small capacitance value for the given circuit. However, with the new Cerafine black series, things are not so simple as to use a larger cap//smaller cap, because each one sounds considerably different. When they work, they're unbeatable and when they don't, save them for a different circuit. I'd have to say that they're worth the trouble, but the amount of trouble can be considerable.
The description of the Blackgate matches exactly the description of the 1uF and 0.47uF Nichicon ES and also the larger Nichicon ES if those are paralleled with a really efficient bypass cap.
Your post is rather awesome, whereby you have applied both power grade and signal grade to the power supply and with good purpose for it.
On the other hand, things (here: opamps) specified (and possibly verified) as having a very low THD tend to sound decent, never "bad"..
Anyway, yes, on a practical ground THD is a parameter of little relevance, especially as a synergic sonic character between opamp & opamp or opamp & DAC chip etc. is so very much more practically influent on the (practical) result than the measurable THD of the entire circuit.
That, BTW, also substantiates my thinking that the LME49723 is preferable to the LM4562 in most cases..
Last edited:
If you refer to the part in Italic thanks but they're not mine but from the site of Audio-Cube (Holland)
But for sure these descriptions apply too to PS decoupling.
I've noted that the timbre and character of caps remains similar in both uses (PS decoupling or signal coupling) during my tests for my optimal selection of components for my Penasa's MyRef.
Also the value could have an important impact on timbre and quality, noted swapping several different PS decoupling Silmics of different value in my SACD player.
By adding/increasing local feedback the OLG is lowered, and THD rises, but the amp sounds the same.
By using active current sources on top of the LTP the distortion is radically reduced, but to my ears the sound seems more distorted and less neutral.
In a discrete amp, I can roll transistors, and I find them sounding different from each other but the THD is exactly the same.
Using a cascoded input stage or a CFP input stage doesn't lower THD but the sound improves. I can see some alterations in the harmonic spectrum though.
Many opamps have very low distortion but still sound different from each other.
The harmonic content probably makes an audible difference, at least that's what I've read in several reports. I haven't done such tests myself.
There simply must be other factors that play a far more important role than THD, but we don't know what to measure or look for.
Hi ClaveFremen,
Thanks for sharing ... however, please note that the Blackgate N 10uF 50V are in the input coupling location that was previously using Nichicon NP ES 10uF 16V. While I think cap rolling is quite straightforward, this recent Op Amp replacement have me worried as some had rightfully warned of more needed details and knowledgement .... e.g. will there be needed change in R and C around the Op Amp when replacement are used??
Just looking at the ecp.cc link on coupling capacitors, perhaps the designer of my amplifier had really gotten the implementation of the Nichicon NP ES caps perfected with the required selection of bypass capacitors etc. and hence better sounding than even the reputed Blackgate N caps which may not work well with these bypass capacitors as the reviewer had also reported:
ecp.cc
I will certainly remove the costly Blackgate N caps and put back the Nichicon ES and on the hindsight since 24/7 for 2 to 3 weeks or 340 to 500 hours should be more than acceptable for any burn-in of capacitors !! Comparatively, recently replacement of Nichicon ES with new ones only took few days to sound very good and the dual OPA627 may just keep the clarity and transparency without sound too bright.
Did not venture to replacement of PS decoupling since it already cost me USD10/each BGN for the 8 input coupling and 4 possibly bypass function.
Phonon340C 10112008013 on Flickr - Photo Sharing!
Regards,
Richard
Thanks for sharing ... however, please note that the Blackgate N 10uF 50V are in the input coupling location that was previously using Nichicon NP ES 10uF 16V. While I think cap rolling is quite straightforward, this recent Op Amp replacement have me worried as some had rightfully warned of more needed details and knowledgement .... e.g. will there be needed change in R and C around the Op Amp when replacement are used??
Just looking at the ecp.cc link on coupling capacitors, perhaps the designer of my amplifier had really gotten the implementation of the Nichicon NP ES caps perfected with the required selection of bypass capacitors etc. and hence better sounding than even the reputed Blackgate N caps which may not work well with these bypass capacitors as the reviewer had also reported:
ecp.cc
I will certainly remove the costly Blackgate N caps and put back the Nichicon ES and on the hindsight since 24/7 for 2 to 3 weeks or 340 to 500 hours should be more than acceptable for any burn-in of capacitors !! Comparatively, recently replacement of Nichicon ES with new ones only took few days to sound very good and the dual OPA627 may just keep the clarity and transparency without sound too bright.
Did not venture to replacement of PS decoupling since it already cost me USD10/each BGN for the 8 input coupling and 4 possibly bypass function.
Phonon340C 10112008013 on Flickr - Photo Sharing!
Regards,
Richard
You're welcome richard!
I've thought that you swapped also PS bypass caps...
Unless there's a particular reason I don't think you need to change R and C around opamps
Nichicons ES are pretty good caps so, if you don't like the BGs sound it should be wise to restore the original caps.
Only BGs costs so much... Silmics and Nichicons costs less than a dollar each...
I don't believe in the "magic" of those for audio caps when used in power supplies.
The Elna Silmic didn't seem anything special to me (in power supply - while in the signal path it's good indeed), as I recall. The Cerafine (as I recall) being more open sounding, enough to allow for removal of power supply regulation. But I'd expect the low ESR 'lytics I use today to behave just the same as the Cerafine
I will consider using boutique caps as power supply caps only when: a) I don't feel 100% happy with Rubycon's low ESR caps (if it ever happens); b) temperature thus long term reliability isn't a concern at all .
The Elna Silmic didn't seem anything special to me (in power supply - while in the signal path it's good indeed), as I recall. The Cerafine (as I recall) being more open sounding, enough to allow for removal of power supply regulation.
But I'd expect the low ESR 'lytics I use today to behave just the same as the Cerafine I will consider using boutique caps as power supply caps only when: a) I don't feel 100% happy with Rubycon's low ESR caps (if it ever happens); b) temperature thus long term reliability isn't a concern at all
.
Last edited:
I don't believe in the "magic" of those for audio caps when used in power supplies.
The Elna Silmic didn't seem anything special to me (in power supply - while in the signal path it's good indeed), as I recall. The Cerafine (as I recall) being more open sounding, enough to allow for removal of power supply regulation.
I will consider using boutique caps as power supply caps only when: a) I don't feel 100% happy with Rubycon's low ESR caps (if it ever happens); b) temperature thus long term reliability isn't a concern at all
No magic at all, Andrea.
PS bypass caps could be considered in the signal path so their character is important.
Silmics compared to normal elcos are quite lower ESR but not enough to be considered low ESR so they're not bad for PS also from the electrical viewpoint.
Rubycon ZL that you use should be a good digital PS cap but for analog decoupling it's not good, IMHO it's a bit harsh with a strange grey midrange...
Panasonic FMs are still a bit harsh but at least their timbre is quite correct
And please stop telling that a cap that costs 0.44$ each is a 'boutique' cap, it's ridiculous...
BTW, when using pretty sterile (color wise) opamps such as the OPA627, I guess that those "sound warming" Cerafine @ Co. caps (in power supplies) must become a strict necessity.
I'm still waiting my OPA1612 samples from TI so I could compare them with OPA627 but to my ears OPA627 is not sterile at all.
Obviously it must be decoupled with at least 47uF caps that are not ZLs, as they impose their grey midrange to the opamp...
Hi ClaveFremen/Andrea,
Just put back 2 original phonon 340C and it was back to musical bliss again.
Somehow, BG N for input coupling and dual OPA627 was way way too bright, harsh and unmusical. I have decided to run-in the modded amps for another 2 weeks of 24/7 to eliminate any run-in effect especially on the BG N and if it still does not work out, I will try one with original Nichicon BP ES with OPA627 and Blackgate with original NJM353 to find out what contribute to the loss in SQ and if I should resign to the original components .... before further soldering/desoldering and damaging the PCB in the process.
On Silmic caps, I previously had one Phonon amp with the original Nichicon FA used in both PS decoupling and entire analog section, including the input coupling, replaced by Silmic II and it was also another expensive disaster ... the amp sounded flat, unmusical, slow and without details despite >100 hour run-in. Guess it all depends on the application, mix and quite a bit of knowledge unlike myself who basically refers to various "best" reviews ... worst still without the schematic to know what is already there that also need to be tweaked. As an example, the bypass with Nichicon NP ES may not be suitable for the BG N which may be not sound good with bypass cap or needing changes in the values.
Of course, I am speculating as this is way beyond my knowledge in EE but guess the original designer had already "voiced" the final design with all these C, R and their values to best performance while I was just doing expensive and potential damaging mods via trial and error with use of the "best" reviewed components in the hope that it would sound better ! Even for the op amp swap, I would think that there are due consideration for Rs as one had advised since current drawn differs and it may also be outright unsuitable if the original voltage is not optimum for the new op amp and needing further mods.
Thanks again so far for the information, comment and advice.
Cheers,
Richard
Just put back 2 original phonon 340C and it was back to musical bliss again.
Somehow, BG N for input coupling and dual OPA627 was way way too bright, harsh and unmusical. I have decided to run-in the modded amps for another 2 weeks of 24/7 to eliminate any run-in effect especially on the BG N and if it still does not work out, I will try one with original Nichicon BP ES with OPA627 and Blackgate with original NJM353 to find out what contribute to the loss in SQ and if I should resign to the original components .... before further soldering/desoldering and damaging the PCB in the process.
On Silmic caps, I previously had one Phonon amp with the original Nichicon FA used in both PS decoupling and entire analog section, including the input coupling, replaced by Silmic II and it was also another expensive disaster ... the amp sounded flat, unmusical, slow and without details despite >100 hour run-in. Guess it all depends on the application, mix and quite a bit of knowledge unlike myself who basically refers to various "best" reviews ... worst still without the schematic to know what is already there that also need to be tweaked. As an example, the bypass with Nichicon NP ES may not be suitable for the BG N which may be not sound good with bypass cap or needing changes in the values.
Of course, I am speculating as this is way beyond my knowledge in EE but guess the original designer had already "voiced" the final design with all these C, R and their values to best performance while I was just doing expensive and potential damaging mods via trial and error with use of the "best" reviewed components in the hope that it would sound better ! Even for the op amp swap, I would think that there are due consideration for Rs as one had advised since current drawn differs and it may also be outright unsuitable if the original voltage is not optimum for the new op amp and needing further mods.
Thanks again so far for the information, comment and advice.
Cheers,
Richard
I mean they would be magical compared to a non-audio low ESR electrolytic bypassed with a good film cap, as I'm used to doing.
I think I checked in their datasheet and their ripple current was pretty low...
I never had such experience of the ZL. Maybe you should've burned them in before judging (or I don't know what)
(a bit like with the Solen caps )Panasonic FM are not open enough sounding, they're mellowed down rather than harsh.
Even just .44$ multiplied by the number required becomes rather snobish, for a electrolytic capacitor. It's just not the kind of part where I'd spend more money, but you're free to think differently.
Oh, well.
My judgment of the OPA627 was formed before I started using the Rubycon ZL (and YXH and ZLH) caps. There were Panasonic FM power caps, where I last used it. It's not just "sad" color wise, but also kind of dry and clynical in the mid-treble. If you try the OPA1611, you'll know what I mean.
Last edited:
Richard,
the 340C looks rather smashing, i've have seen quite a few flash by at Japanese forums.
Do you have something remotely equivalent to a schematic of this amp, by chance ?
I'm sorry Andrea but you're wrong, let's consider a standard value of 100uF 50V, here it is the Rated Ripple Current from their respective datasheets:
Nichicon VR 260 mA
Rubycon YK 279 mA
Elna Cerafine 310 mA
Elna Silmic II 380 mA
Elna RJH (Low ESR) 533 mA
Panasonic FC (Low ESR) 615 mA
Rubycon ZL (Low ESR) 724 mA
Panasonic FM (Low ESR) 950 mA
As you can see Silmics have 40% more Ripple Current than good quality normal elcos but not enogh to be like low ESR, as I've said...
And FMs are the better than ZLs...
I never had such experience of the ZL. Maybe you should've burned them in before judging (or I don't know what)
They stayed in my SACD player and my receiver for two years, is it enough for consider them burned-ind?
Panasonic FM are not open enough sounding, they're mellowed down rather than harsh.
FMs not open?Thanks for letting me choose my caps...
But we're talking about what is 'boutique' and Silmics are normal industrial caps, a component to be called 'boutique' must have an insane cost and normally is handmade by virgin girls...
You could say that they costs more than you like for an industrial cap but no way they're 'boutique'
We must wait until April...
For now I've:
OPA2107P, OPA627AU, OPA827, OPA2132UA, OPA2134P, OPA2211, OPA1642, LT1469, JRC 2068, JRC5532DD, Signetics NE5532N, TI NE5532P, LM6172, LM4562
but I still have to try the underlined ones...
Really the Rubycon ZL/ZLH don't impose a grey midrange on anything, or I'd have hated them.
IMHO they do, pulling them out of my gear was a revelation!
As for burn-in read my previous answer.
Last edited: