Hi fellow diy-ers! After a long diy break i came back to my beloved MyRef and i read the last developments and i want to upgrade my boards with the latest BOM and upgrade to OPA827.
While i'm updating bit by bit my Mouser cart i am confused whether to choose OPA827AID or OPA827AIDR or OPA827AIDG4. I can't find the explanation of those suffixes.
Noticed semnificative changes in R101/104/201/204. Really makes a difference?
Thanks and stay safe from this new coronavirus bug!
While i'm updating bit by bit my Mouser cart i am confused whether to choose OPA827AID or OPA827AIDR or OPA827AIDG4. I can't find the explanation of those suffixes.
Noticed semnificative changes in R101/104/201/204. Really makes a difference?
Thanks and stay safe from this new coronavirus bug!
George will be the ultimate knowledge base, but from what I know, they are the same op-amp, just different packaging. I bought the AID, because it was available at the time and a tube is a convenient package. But I think you could buy any of those.
In the US at Mouser, only the AID is available as an order of 1.
In the US at Mouser, only the AID is available as an order of 1.
Hi!
OPA827 AID or OPA827 AIDR is ok. Be careful, the 'g' versions are small package, not SOIC.
For the resistors, Dario would be the authority. But if you have the BOM, (with Dale), those should be very fine.
Ciao, George
OPA827 AID or OPA827 AIDR is ok. Be careful, the 'g' versions are small package, not SOIC.
For the resistors, Dario would be the authority. But if you have the BOM, (with Dale), those should be very fine.
Ciao, George
I could have easily gotten caught out on that one. They are showing a GKR and a GKT in VSSOP size and a G4 and an RG4 in SOIC, along with the AID plain and R SOIC. Its way down in the datasheet and I totally missed it.
As for the coronavirus, I hope everyone is safe. I am planning on using the stay home time to move an audio project forward.
All the Best,
Jac
As for the coronavirus, I hope everyone is safe. I am planning on using the stay home time to move an audio project forward.
All the Best,
Jac
Atupi: I ordered #595-OPA827AID from Mouser and it fit/worked wonderfully. The 5pF silver mica cap that was soldered across pins 2to6 was from Digikey...CD15CD050D03F. My thanks to both George and Jac for their inputs on that build. I’m forgetting someone else, sorry.
I did not like 827 (C9 shorted or not) . I tried OPA627 (C9 shorted or not) after that (didn’t wow me) and finally settled on ADA4627 with C9 shorted (for now). All my variations are compared to Dario’s MY_Ref Evo A with audiophile build parts (I still love that build).
Regards,
Pete
I did not like 827 (C9 shorted or not) . I tried OPA627 (C9 shorted or not) after that (didn’t wow me) and finally settled on ADA4627 with C9 shorted (for now). All my variations are compared to Dario’s MY_Ref Evo A with audiophile build parts (I still love that build).
Regards,
Pete
Pete,
I am happy to see that experience is starting to accumulate.. Thank You.
You had also mentioned that You might like better in original form, with the LM318 in place.
If you feel adventurous, maybe the 828 worth a try, too.
I have just recently (thanks to Giacinto.. 🙂) finished the pair of FE that I started experiment with when Jac contacted me about the stability problems in his FE with 828.
After that I continued to use my EVO (habits..)
But now that I completed that pair equipped with upper- medium parts (industrial BOM, with styroflex/armtrans /PPS capacitors, VishayS102 in the bridge and Holco/ takman rey in the feedback) -
I did like what I heard.. 🙂 and it is basically a poor parent of the amp of Jac.
And.. Tweak time is in arrival... That is, it had just crashed on us here.
Greetings, George
I am happy to see that experience is starting to accumulate.. Thank You.
You had also mentioned that You might like better in original form, with the LM318 in place.
If you feel adventurous, maybe the 828 worth a try, too.
I have just recently (thanks to Giacinto.. 🙂) finished the pair of FE that I started experiment with when Jac contacted me about the stability problems in his FE with 828.
After that I continued to use my EVO (habits..)
But now that I completed that pair equipped with upper- medium parts (industrial BOM, with styroflex/armtrans /PPS capacitors, VishayS102 in the bridge and Holco/ takman rey in the feedback) -
I did like what I heard.. 🙂 and it is basically a poor parent of the amp of Jac.
And.. Tweak time is in arrival... That is, it had just crashed on us here.
Greetings, George
Thanks for all your infos!
Why did you installed the 5pf capacitor on OPA827 pins? You had oscillations without it?
Could you describe what you did not liked on 827 and liked on ADA4627?
All the best,
Adrian
Because I had suggested..
No, it is not oscillating without. In fact, some days ago I had suggested to Pete also to try to remove it.
The extra cap gives more phase margin, which somehow brings a more 'controlled' manner in the sound.
In change for a less vivid, say, spectacular presentation without.
(in my experience, my system(s)..)
Also the feedback capacitor type is notable: i have found the mica capacitors bring more open, less constrained sound, while C0G caps are less spectacular, more 'simple' sounding, uhm could call it 'grey'.. But more neutral.
What I'm trying to say here is for Pete - there are more ways to 'deviate' the sound of the system, maybe worth trying. I did not understand yet his quest, the type of sound, although the 4627 is a hint..
Ciao, George
No, it is not oscillating without. In fact, some days ago I had suggested to Pete also to try to remove it.
The extra cap gives more phase margin, which somehow brings a more 'controlled' manner in the sound.
In change for a less vivid, say, spectacular presentation without.
(in my experience, my system(s)..)
Also the feedback capacitor type is notable: i have found the mica capacitors bring more open, less constrained sound, while C0G caps are less spectacular, more 'simple' sounding, uhm could call it 'grey'.. But more neutral.
What I'm trying to say here is for Pete - there are more ways to 'deviate' the sound of the system, maybe worth trying. I did not understand yet his quest, the type of sound, although the 4627 is a hint..
Ciao, George
Thanks for your advices George! I my head was the idea that i need to use the extracap only when oscillating. I will install it anyway in this case.
when you are speaking about feedback cap you mean C32?
when you are speaking about feedback cap you mean C32?
No, I was talking exclusively of that 'noname' extra local feedback capacitor, on Pin2-Pin6 of the opamp..
Obviously also C32 has it's characteristics, but that is more than well covered in this thread (threads!!).
Ciao, George
Obviously also C32 has it's characteristics, but that is more than well covered in this thread (threads!!).
Ciao, George
Atupi: As George stated about the mica cap across 2&6, I sought his wisdom on the 827 build and he suggested it. Who was I to argue 😉? So I did not run the chip w/o the cap on it so I can’t make a comment about oscillations. Sorry. OPA827 was cleaner sounding and articulated the sound a bit better than LM318 but, at least on my system I lost the beautiful soundstage & airy quality that the original Dario build had (I’ve never experimented with the original build so I could go back and forth to compare). I let the 827 build burn in for about 60 hrs of listening time before I switched out to the OPA627 with a 3.3pF silver mica across 2&6 (again, a George suggestion but I can’t make comment about oscillations since I never tried 627 w/o it). To my ears, it did not articulate the music like 827 did and the soundstage was compressed compared to the original Dario build.
None of them sounded “bad” but didn’t have the ethereal quality of the original LM318 build until I tried ADA4627 (5pF silver mica across 2&6) with C9 shorted (I’ll try it with C9 in place as well). I’m currently running the 4627 build for awhile. The bottom frequency response is not as sharp as with the LM318 build but solid none-the-less. I made the comment to George I’m beginning to think preferences for any one of these chips you use for your builds will depend highly on your speakers, DAC units you use, the age of your ears, choices of evaluating music, etc, etc. I remain humbly bowing to Dario, Jac and George for their knowledge & experimentation experience.
Cheers, Pete
None of them sounded “bad” but didn’t have the ethereal quality of the original LM318 build until I tried ADA4627 (5pF silver mica across 2&6) with C9 shorted (I’ll try it with C9 in place as well). I’m currently running the 4627 build for awhile. The bottom frequency response is not as sharp as with the LM318 build but solid none-the-less. I made the comment to George I’m beginning to think preferences for any one of these chips you use for your builds will depend highly on your speakers, DAC units you use, the age of your ears, choices of evaluating music, etc, etc. I remain humbly bowing to Dario, Jac and George for their knowledge & experimentation experience.
Cheers, Pete
Last edited:
Absolutely, a BIG one too!
You will gain a much more detailed sound, the Caddocks in the original BOM hid a not small amount of details.
We're trying to do it at our best... we're stuck at home apart for food procurement, health reasons, going to work if smart working is not possibile.
More and more countries in Europe are going to adopt policies similar to Italian ones, social distancing is the only way to go.
It's hard, though.
Wish you the best
Dario,
I know it is very trying in your area. Hope your family and friends have minimal impact.
Suspect the rest of us will feel some of your pain soon.
Dario,
I know it is very trying in your area. Hope your family and friends have minimal impact.
Suspect the rest of us will feel some of your pain soon.
My family and friends are doing well so far, thanks 🙂
The sooner governments will take strict social distancing policies the less will be the pain.
Well said. The only right answer is the one you choose for yourself. But we keep trying new stuff, just in case. That's DIY.
I have my OPA828 FE's in my main system and my 4 channel LM318 FE's in my project room. Both sound great and match the system they are with very well.
Dario and everyone,
Do stay safe. We are quickly catching up with Italy with social distancing and lockdowns here is the US. I'm glad I live in the country, so I can walk around outside without other people nearby. I am also grateful for these projects to keep my mind occupied. I can't imagine some of my friends without hobbies, sports, or travel. They must be going crazy.
All the Best,
Jac
Jac & George:
Regarding 828, is it a drop in replacement for 827/627/4627 in my Evo.A? Will I have to use a 100pF cap at C32 as I did with 827? 5pF across 2&6 again as a precaution? What subjective opinions can you impart about 828 compared to the stable of other op-amps discussed in this thread with regard to its musicality?
And, a question I’ve been meaning to ask about “oscillations” mentioned with regard to op-amp rolling (if that’s the acceptable lingo)...I don’t have an oscilloscope so would I be able to actually hear these oscillations through my system as I am listening to music?
Finally, to all of you in Italy, I’ll join Jac in sending our heartfelt sympathies to all those affected there. I’m on the East Coast of the USA and we are hoping that everyone takes this virus’ s spread seriously.
Regards, Pete
Regarding 828, is it a drop in replacement for 827/627/4627 in my Evo.A? Will I have to use a 100pF cap at C32 as I did with 827? 5pF across 2&6 again as a precaution? What subjective opinions can you impart about 828 compared to the stable of other op-amps discussed in this thread with regard to its musicality?
And, a question I’ve been meaning to ask about “oscillations” mentioned with regard to op-amp rolling (if that’s the acceptable lingo)...I don’t have an oscilloscope so would I be able to actually hear these oscillations through my system as I am listening to music?
Finally, to all of you in Italy, I’ll join Jac in sending our heartfelt sympathies to all those affected there. I’m on the East Coast of the USA and we are hoping that everyone takes this virus’ s spread seriously.
Regards, Pete
Pete,
First the factual part:
Configuration with 828:
Cap Pin2-pin6: 5pF
C32: 68pF ( i tried from 60pF to 75pF, like two pieces of 150pF in series)
These are a must here, not optional.
And I would say, that with the exception of 827, all other places that 5pF local feedback cap is needed.
Serious instability results in excess heat in the output devices (3886), distortion, 'thin sound', very probably also the protection kicks in (depends on how much of the signal gets rectified).
Slight amounts are more tricky to understand, especially if it's not a constant status, but kicks in occasionally.
Heat in the output stages is always a good indicator, even if it is not elevated. Obviously, with high volume you get heat anyway..
But. Due to the system general configuration, like the clipping protection, which does not allow the signal to reach above +_28Vpeak, (56Vpp) -- the myref topology is usually very much in the safe zone of operation.
George
First the factual part:
Configuration with 828:
Cap Pin2-pin6: 5pF
C32: 68pF ( i tried from 60pF to 75pF, like two pieces of 150pF in series)
These are a must here, not optional.
And I would say, that with the exception of 827, all other places that 5pF local feedback cap is needed.
Serious instability results in excess heat in the output devices (3886), distortion, 'thin sound', very probably also the protection kicks in (depends on how much of the signal gets rectified).
Slight amounts are more tricky to understand, especially if it's not a constant status, but kicks in occasionally.
Heat in the output stages is always a good indicator, even if it is not elevated. Obviously, with high volume you get heat anyway..
But. Due to the system general configuration, like the clipping protection, which does not allow the signal to reach above +_28Vpeak, (56Vpp) -- the myref topology is usually very much in the safe zone of operation.
George
Hi Pete,
I will let George speak to the sound. And, of course, he as the most knowledge on stability.
But I will let you know that the 828 is a much faster op-amp, so oscillation is a bigger risk. Per George's simulation and experiment, I needed and used both the 5 pF across pins 2 and 6 and C32 needs to change to 68 pF. Those changes worked for me. As George has said, the 828 can work well, but needs a bit of bravery.
I found that an industrial polystyrene at C32 was the best and most neutral sounding choice for 68 pF. I tried silver mica and a COG ceramic and they both sounded surprisingly different and worse (kind of flat). At 68 pF, there aren't many choices.
I tried 4627, then 627, back to LM318, then a year later 828. So my ability to compare is compromised by time. I did like all 3 jFet op-amps in my system, but that's no guarantee for you. And I still love the LM318 in my other system.
As far as oscillation, its not perfect, but if you measure the output with the input shorted and you have significant DC and AC on your meter, then you may be oscillating. BTW, these USB oscilloscopes are pretty good for less than $100.
One last thought about rolling jFet op-amps. Make sure you are using a no-clean flux for both the solder and sticky flux. Its not totally proven, but it looks like flux can stay under the op-amp chip and allow conduction between pins. I struggled with DC offset until cleaning up this problem. This could happen with the LM318, but the jFet op-amps seem more sensitive. Of course, without C9, the op-amp is at much higher gain, so any problem is amplified.
Jac
PS Naturally George was typing at the same time I was. I bow to the expert.
I will let George speak to the sound. And, of course, he as the most knowledge on stability.
But I will let you know that the 828 is a much faster op-amp, so oscillation is a bigger risk. Per George's simulation and experiment, I needed and used both the 5 pF across pins 2 and 6 and C32 needs to change to 68 pF. Those changes worked for me. As George has said, the 828 can work well, but needs a bit of bravery.
I found that an industrial polystyrene at C32 was the best and most neutral sounding choice for 68 pF. I tried silver mica and a COG ceramic and they both sounded surprisingly different and worse (kind of flat). At 68 pF, there aren't many choices.
I tried 4627, then 627, back to LM318, then a year later 828. So my ability to compare is compromised by time. I did like all 3 jFet op-amps in my system, but that's no guarantee for you. And I still love the LM318 in my other system.
As far as oscillation, its not perfect, but if you measure the output with the input shorted and you have significant DC and AC on your meter, then you may be oscillating. BTW, these USB oscilloscopes are pretty good for less than $100.
One last thought about rolling jFet op-amps. Make sure you are using a no-clean flux for both the solder and sticky flux. Its not totally proven, but it looks like flux can stay under the op-amp chip and allow conduction between pins. I struggled with DC offset until cleaning up this problem. This could happen with the LM318, but the jFet op-amps seem more sensitive. Of course, without C9, the op-amp is at much higher gain, so any problem is amplified.
Jac
PS Naturally George was typing at the same time I was. I bow to the expert.