[Les Lammers]

Emac,

I have never swapped opamps and I agree that they may make more of a difference than caps. If the 2107 is a bit warmer than the stock unit it may be the ticket. I am not a digital designer and this is uncharted ocean for me. 

[emac]

Yes, the 2107 is warmer than the stock opamp.  But it also changes the soundstage, making the sound less recessed and more forward. 

[Les Lammers]

I bought the 2107 and will give it a go and post the results. *I* am not sure if I want the soundstage more forward. I don't care for an in your face presentation. Just personal preference. But it may just be fine. Do opamps have a burn in? What opamp was in your Constantine?

[emac]

I bought the 2107 and will give it a go and post the results. *I* am not sure if I want the soundstage more forward. I don't care for an in your face presentation. Just personal preference. But it may just be fine. Do opamps have a burn in? What opamp was in your Constantine?

The opamp was the AD826, which really wasn't well suited for the Constantine.  As for burn in, the differences would be apparent pretty much off the bat.  I'd give it an hour or two to settle in, but that's all I've found from my own experiences.  If anyone else has a different opinion, please share it.

[Les Lammers]

Earlier in this thread Legarman posted about changing the Paridisea opamp to the LM4562. Apparently MHDT paid attention:

I bought this DAC six months ago when someone came at home with his homebrew non oversampling Dac and when I read Scott Faller report. 

I really like this dac but in stock form, it lacks a little definition and soundstaging.  I replaced the tube with a 396A.  I began to draw the circuit and took some measuremnents.  V+ and V- for the OPA2604 is +17V and -17V.  This means that the op amp is feed with more voltage that specified by the company.  First, I don't like the OPA2606 very much.  It is a dark sounding chip and lacks definition.  Finally, I decided to place a 8 pin socket at the place of the SOIC chip installed in the Paradisea.  I used a socket because I wanted to try many op amps.  First with +17 and -17V, these supply voltages are too high.  The V+ and V- come from two TL431 shunt regulators that feed the tube cathode follower.  Yes, you read correctly, the cathode follower has a 100 ohm for RP and another 100 ohms for RK. only +19V feeds the 100 ohm RP and only -19V feeds 100 ohms RK.  This means that the tube operates with only 38V (I made a drawing  but can't insert it with my message).  From the +19V and -19V, there's 100 ohm resistors for decoupling and lower the voltage to the OPA 2604.  The op amp is not decoupled with any capacitors at all.  I replaced the two 100 ohms feeding the op amp with 330 ohms resistors and got _+14.5V and -14.5V.  Near the op amp, I placed from V+ and gnd and V- and gnd 10mfd tantalum caps paralled with .01 mfd  ceramic caps.  These caps are placed aside the op amps.  (I took some picts).  Here are the op amps tried:  AD826 = too much high mid, too hard.  Two OPA627 placed on a Browndog adaptor = Mellow sound, very good soundstage, too much laidback sounding for my taste but WAY better than OPA2604. The surprise came from the newcomer National LM4562 specially designed for audio.  When I got samples from National this new chip was on market since only two days.  Soundstage and image are very good.  Definition is excellent.  At first there was not too much bass but things were correct after about 100 hours of playing time. 

Placing this op amp in the paradisea with good decoupling caps across it means a new dac.  BIG LEAGUE name dacs simply can't stand with the Paradisea about sound quality when modified.  More on that later

Legarem

And a reply:

Legarman,

Many thanks for your generous contribution to AC. An auspicious start!

For those seeking a simpler, inexpensive alternative to 'brown dogged' OPA627s, I'd suggest an OPA2107 (surprise, surprise!) which is a dual, available in both SOIC8 and DIP8. For those who like the Burr Brown warm sound, this chip is about as close to the OPA627 in a drop in dual as one can find. It might be a good alternative to the AD826 that comes in the Constantine. While I subjectively like the BB sound, I've always found the OPA2604 gritty, unresolving, unextended and closed in and never found a situation where the OPA2107 wasn't a substantial upgrade. (given the relative prices I'd expect nothing less)

Like the LM4562, the OPA2107 are relatively low speed, low bandwidth (in comparison to many AD chips), but nevertheless wonderful opamps, and unlikely to oscillate  or be otherwise problematic when swapped with most other opamps. The LM4562 is an appealing alternative to AD chips for those seeking less warmth than typical in BB opamps. ( I've presently one cooking in the center position of my Zhaolu 2.0)

I believe the easiest way to desolder a SOIC8 chip, lacking access to appropriate desoldering equipment, is to use a safety pin and slide the point between the body and the legs as you desolder. Slide the pin point further in as you desolder each pin. When you've lifted the pins on one side, repeat on the other side.

Regards,
Paul

The opamp seems to make a difference in what caps, if any, sound better, Robert Brown installed the OIMP's in the original Paradisea and loves them.

[emac]

Placing this op amp in the paradisea with good decoupling caps across it means a new dac.  BIG LEAGUE name dacs simply can't stand with the Paradisea about sound quality when modified.  More on that later

We'll be testing this idea soon enough.  My modded Constantine will be going up against some well regarded, big name DACs (e.g. Benchmark, modded DAC-60) coming up next weekend.  The list of DACs hasn't been finalized, so I'll hold on listing them for now.  But we'll post the results once we're done. 

[Les Lammers]

Placing this op amp, National LM4562 , in the paradisea with good decoupling caps across it means a new dac.  BIG LEAGUE name dacs simply can't stand with the Paradisea about sound quality when modified.  More on that later

We'll be testing this idea soon enough.  My modded Constantine will be going up against some well regarded, big name DACs (e.g. Benchmark, modded DAC-60) coming up next weekend.  The list of DACs hasn't been finalized, so I'll hold on listing them for now.  But we'll post the results once we're done. 

Emac, I added the National LM4562 to your quote for the sake of clarification. I am going to let the DAC run 24/7 with a signal for another week while I am waiting for the 2107. I may also give AN Copper caps a go with the LM4562 too. It just may be that the LM4562 and OIMP's are not an optimum combination in my system and room. Even though it is regarded as a no no, I may bypass the OIMP's with some small value Vitamin Q's that I have. I can always snip them off. 

Also, the combination of the LM4562 and OIMP's may have raised the bar high enough to expose the weakness of the PS caps. I may try an Isoclean fuse too. Thoughts? 

[emac]

Emac, I added the National LM4562 to your quote for the sake of clarification. I am going to let the DAC run 24/7 with a signal for another week while I am waiting for the 2107. I may also give AN Copper caps a go with the LM4562 too. It just may be that the LM4562 and OIMP's are not an optimum combination in my system and room. Even though it is regarded as a no no, I may bypass the OIMP's with some small value Vitamin Q's that I have. I can always snip them off. 

Also, the combination of the LM4562 and OIMP's may have raised the bar high enough to expose the weakness of the PS caps. I may try an Isoclean fuse too. Thoughts? 

There's nothing wrong with bypassing caps.  It's not to everyone's liking and it doesn't help in all circumstances.  But it's worth a shot if you're interested.  I've got mine bypassed, and I like the results. 

As for the fuse, if you check back a page, I swapped out the stock slow blow fuse with a Radio Shack slow blow of roughly the same value.  I won't be going back.  A cleaner and more detailed sound for $3.  Can't go wrong with that.  So, the Isoclean fuses might be worth a shot, though I looked at that a little, and it seems like the Hi-fi tuning fuses are typically better liked.  $5 cost difference between them. 

[Les Lammers]

I have the caps and it is a 5 minute job. Isoclean v. whatever...I dunno. I wudda never thunk about fuses but ya never know. I have a Rat Shack close by so I may spend $3 there. Digital...no sources in general can make you nuts. Analogue is no piece of cake either. No pain, no gain.

[Les Lammers]

Emac, I added the National LM4562 to your quote for the sake of clarification. I am going to let the DAC run 24/7 with a signal for another week while I am waiting for the 2107. I may also give AN Copper caps a go with the LM4562 too. It just may be that the LM4562 and OIMP's are not an optimum combination in my system and room. Even though it is regarded as a no no, I may bypass the OIMP's with some small value Vitamin Q's that I have. I can always snip them off. 

Also, the combination of the LM4562 and OIMP's may have raised the bar high enough to expose the weakness of the PS caps. I may try an Isoclean fuse too. Thoughts? 

There's nothing wrong with bypassing caps.  It's not to everyone's liking and it doesn't help in all circumstances.  But it's worth a shot if you're interested.  I've got mine bypassed, and I like the results. 

As for the fuse, if you check back a page, I swapped out the stock slow blow fuse with a Radio Shack slow blow of roughly the same value.  I won't be going back.  A cleaner and more detailed sound for $3.  Can't go wrong with that.  So, the Isoclean fuses might be worth a shot, though I looked at that a little, and it seems like the Hi-fi tuning fuses are typically better liked.  $5 cost difference between them. 

http://209.85.135.104/translate_c?hl=en&langpair=de%7Cen&u=http://cgi.ebay.de/ahp-Feinsicherungen-II-neue-Version-5x20mm_W0QQitemZ360027010233QQihZ023QQcategoryZ141226QQcmdZViewItem&_trksid=p1742.m153.l1262

The link is to a German seller on ebay. I bought 3 fuses to try. A Furutech and one each of the AHP copper and gold. I dunno how he sells the Furutech for $20. I have seen then for $50 here. The furutech was 20 Euros about $32. I got some Radio Shack fuses and they are better than what was in the unit. They seem to be directional. I cannot believe I typed that. 

[Les Lammers]

I by passed the OIMP's with .033 uf Vitamin Q's and out of the gate the sound is headed in the right direction. I would not recommend the OIMP's for use in the Paradisea +. However, they are excellent caps and may be just right in the original Paradisea. The LM4562 opamp and the OIMP's are too much of a good thing.

[t-head]

Thanks, Les,

You just saved me some $$$...

Richard

[Les Lammers]

Hi Richard,

The P+ has a different opamp and transformer than the original and that may make the difference regarding caps. I did order some Obbligatto oil caps from:

http://www.diyhifisupply.com/diyhs_electronic_parts.htm They are inexpensive a pretty well thought of.

I also have some 2.0uf Vitamin Q's coming. For $10 each they are worth trying. This is a nice little DAC and IMHO experimenting with caps etc. is worthwhile. The stock caps bypassed with a Russian teflon may be good too.

Les

[Les Lammers]

  I think the biggest improvement to the DAC was the Chimera digital cable.  I bought it after I installed the OIMP's. I was using a Stereovox HDVX and could not go back to it after hearing the Chimera. I put the stock caps back and dammit they sound pretty good with the Chimera cable and a Bendix 2c51.  The tonal balance is back and the sound is extremely detailed. YMMV as always but the cable made a bigger difference than the OIMP's. I will post after the VQ's and Obbligatto's are tried.

[t-head]

Les,

I await your judgement... FWIW, the change to coax input with the Trends UD10 and a Pulsar IC from VH Audio (rather than native USB input) was a big change...

Richard

[Les Lammers]

Richard,

In the end all this is an experiment and what matters is how it sounds in your system. For someone with different electronics and speakers the OIMP's may be the ticket. It is amazing as well as frustrating how changing one part can tip the balance.

Les

[t-head]

Les,

I hear ya...ain't this a wonderful hobby...

Richard

[Les Lammers]

I ordered 2 sheets of ERS paper yesterday. I think the glare I am hearing, and is still there but not as noticeable with the stock caps, was accentuated by the OIMP's. Will experiment with it on the transport and DAC.

[emac]

I ordered 2 sheets of ERS paper yesterday. I think the glare I am hearing, and is still there but not as noticeable with the stock caps, was accentuated by the OIMP's. Will experiment with it on the transport and DAC.

I found that the ERS worked well in 2 spots, which is similar to the results that I got in my amps.  First, put the ERS on top of any chips.  I didn't bother with the opamp because it didn't make any difference and it's pretty small to begin with.  The 2nd, and more important area, is around the transformer.  Nice effect here, similar to what I've found with my amps.  Be careful though.  There's not a whole lot of room around the transformer, so the ERS will be close to some of the parts.  And you don't want the ERS to be touching any of the parts in the circuit because it is flammable.  However, if you're careful with your placement and put tape over the edges of the ERS (which is the flammable part) it should be okay.  That's what I've done, and haven't had a problem.  But it's your DAC, and your risk. 

[Les Lammers]

Emac,

I bought the adhesive sheets and will give it a go. I will be sure it is placed properly B4 I place it permanently and plan to put electrical tape on the edges of the paper. I am also going to put it inside the CD player. Sometimes you do not notice an issue until you change something. I am getting the same glare without the DAC and it is still there with a different CD player. It is also possible that a new source of EMI/RFI is present in my area. Sometimes I think about getting an Almarro 205, small speakers and forgetting about the chase. 

Les