Thanks, Jack. I could not find that, but now it's bookmarked!
One question: why must compensation parts C34, R43 be removed? The 827 pins 1, 5, 8 have no internal connection, so those comp parts are isolated, sitting idle. Can they affect the circuit in any way?
Okay, one more: could C34 pads be populated with the stabilizing cap (if needed) with one end jumpered to pin 2?
Peace,
Tom E
One question: why must compensation parts C34, R43 be removed? The 827 pins 1, 5, 8 have no internal connection, so those comp parts are isolated, sitting idle. Can they affect the circuit in any way?
Okay, one more: could C34 pads be populated with the stabilizing cap (if needed) with one end jumpered to pin 2?
Peace,
Tom E
Yes. Jac had done exactly that.
Tom, unfortunately I have an ailing internet access right now. But it would be nice to keep contact in these days.
I prefer remove everything - - I do not trust stray impedances in the package, even if basically You are right.
Do you already have the chip? It seems that You are mostly ready to go..
The opamp change principally aimed to bring a better control, let's call it like this, and always IMHO.
The C9 'affair' is a gift of the company..
One note: be careful with that 'power supply' steal- trick.
That is not compatible..
Ciao, George
Tom, unfortunately I have an ailing internet access right now. But it would be nice to keep contact in these days.
I prefer remove everything - - I do not trust stray impedances in the package, even if basically You are right.
Do you already have the chip? It seems that You are mostly ready to go..
The opamp change principally aimed to bring a better control, let's call it like this, and always IMHO.
The C9 'affair' is a gift of the company..
One note: be careful with that 'power supply' steal- trick.
That is not compatible..
Ciao, George
Tom,
I haven't looked closely at the 827 pins, but it sounds like what you suggest would work. I hope George will comment as he has done most of the work on these opamps.
If you need a stabilizing cap, what you suggest should be fine. Just keep the jumper as short as possible. I am assuming you were planning and SMD cap for this purpose and that makes a lot of sense. I actually did it the hard way with a radial cap, one end soldered directly on the pin 2 and the other end on the pad that is right next to pin 6.
I haven't looked closely at the 827 pins, but it sounds like what you suggest would work. I hope George will comment as he has done most of the work on these opamps.
If you need a stabilizing cap, what you suggest should be fine. Just keep the jumper as short as possible. I am assuming you were planning and SMD cap for this purpose and that makes a lot of sense. I actually did it the hard way with a radial cap, one end soldered directly on the pin 2 and the other end on the pad that is right next to pin 6.
OK, I finally got to the part where you guys discussed input opamp mods.
It happens that there are grade A and (better) grade B, which is what you used I think.
What about ADA4637? Did you try it too? The higher slew rate could mean stability issues? Speaking of which, even if it worked, the myref would be slew-limited by the LM3886, correct?
Did you also try mod686 opa of choice, lme49710 IIRC?
I'll keep reading anyway.
Looking at the measures you took this is the most promising IMO.
It happens that there are grade A and (better) grade B, which is what you used I think.
What about ADA4637? Did you try it too? The higher slew rate could mean stability issues? Speaking of which, even if it worked, the myref would be slew-limited by the LM3886, correct?
Did you also try mod686 opa of choice, lme49710 IIRC?
I'll keep reading anyway.
Absolutely, being unable to do to on the FE, I've made this experience on DACs.
Apples to oranges. Sorry, Dario.
I still have about 4m of it, I used on the MF on my previous speakers, driven by the firstwatt F3. I think it's very good, cheap and never had a problem with it.
I think 9m is too much for unshielded speaker cable, though.
@Panelhead:
Upon George's suggestion and some tests, lately we've completely removed the DC servo from our own my_Evos!
As it turns out, indeed it sounds better without it.
(N.B.: of course we are also using JFET-input OpAmps -in my case ADA4627- instead of the LM318, otherwise the servo could not be removed).
Thus, currently our updated my_Evos are really a lot similar to Dario's FEs (with George's mods).
The main (and almost only really significant) difference which remains between the Evo and the FE is the possibility of doubling the current pump, which does exists in the Evo but not in the FE.
In our experiences, the "full-Evo" (with double c.pump) does indeed have some notable advantages -also sound-wise- WRT a single-pump version. At least when paired with speakers which shows lowerish impedances. With "easy load" speakers that have a pretty flat impedance curve, with modest phase shift and which never goes much below 8 Ohm, perhaps the difference might be less evident.
Thus, if you have an Evo board, I'd use that one. In "full" configuration (4*3886, etc), with all George mods. In that configuration and using the right BOM, it's SQ will rival that of the very best amps on the market, no matter the price!
My guess would have been the same about how it sounded.
I'd like to sum up a few pointers:
1) dc servos are not trasparent as their designers wants us to believe (and believe themselves).
2) An extremely good (and expensive) ac-coupling cap could sound better than a servo
3) Modestly priced ada4627 and no c9 is a pretty cheap option to try (with dmm and scope ready)
I wish I could listen to a full evo in the future.
Reason why I'm sticking with (lossless) volume attenuation in the source.
The only issue i see with the myref is its high gain, and reducing it to the point I want will most likely make it unstable.
So, probably in this case a tvc or a high quality (vishay naked and co) stepped attenuator is the best option.
Probably very close to the limit.
Weaknesses are only 2 for me: too high gain and too low slew rate.
Neither of which are easily fixable. Let's just enjoy our chipamps
Sorry, Guido, I didn't like this absolute statement on a decontextualized quote.
In 'absolute' terms comparing opamps in different circuits could give different results and I've clearly stated that my comments could or could not apply.
Nevertheless it's still an interesting test that could give some preliminary hints on how they will contribute to the overall sound with their sound color.
While far from being an absolute truth, in my experience opamps tend to maintain their sound signature in different circuits.
Maybe that sounded a little strong. But think that thay are both fruits
It is my opinion that trying an opamp in a dac I/V can give only a general assestment about its qualities, but that it needs to be put in its intended circuit (i.e. in place of the lm318) to know exactly how it will sound. Some others shares the same opinion.
After all you have done extensive A/B comparisons with passive components previously.
I still don't know (at page 330) if and which opa you have personally tried in the myref_fe.
It is my opinion that trying an opamp in a dac I/V can give only a general assestment about its qualities, but that it needs to be put in its intended circuit (i.e. in place of the lm318) to know exactly how it will sound. Some others shares the same opinion.
After all you have done extensive A/B comparisons with passive components previously.
I still don't know (at page 330) if and which opa you have personally tried in the myref_fe.
The most interesting approach that I have seen to this problem is the Schiit Ragnarok. Quoting their description;
"Nothing In the Signal Path Except Signal
Ragnarok is an "intelligent" amplifier, using a microprocessor to oversee every aspect of its operation—from quiescent bias, to DC offset, to complete fault protection. It’s also designed so that any major problem mutes all the outputs and reverts the amp to its low-gain, zero volume mode. This means we can dispense with coupling caps and DC servos entirely, for a gain stage that has nothing in the signal path—except for your music."
From what I understand, they have an algorithm monitors the signal and tries to decide what is music and controls the bias and DC offset that go with that signal. When Stereophile tried to test it, the test signals were not well received by the algorithm because they aren't as transient as music, therefore, the test showed high distortion. Schiit had to explain what was going on to Stereophile because nobody thought it sounded distorted.
I understand the most recent versions of the algorithm are more forgiving to test signals.
Unfortunately for us, the expertise and manpower to create such a control is probably beyond individuals.
Jac
Having hung around this thread with Dario since 2012, I am willing to be he has tried the OPA827 and ADA4627. The only question is if he tried the opa627 in the FE.
Jac
This is another thing I want to try (worth resurrecting). Since i have a 50R dac source, it's win-win.
In the past three weeks i have read all threads about myref from 2012, except the old group buys. Til page 330 of this one (340 now).
Let's see what he replies.
Oh, it's your design, gz!
Always liked this chip, and I have one of course, not your design though.
this is my DAC ....and unfortunately the author is always me.
my nick in that forum is "bigtube". on diyaudio they are "sontero".
I can safely say that the MyRef -FE with the mods of George reaches a very high level and the combination with my DAC is very happy.
You didn't understand at all, I've made my tests on the LPF section not on IV... The AK4495 has voltage out.
I would have never did such a comparison, I/V it's a completely different circuit with so different requirements...
You listened to my work at your home, you appreciated and loved it.
It's the results of years of work, study, refinement, collaboration of nice and competent people like LinuxGuru and JosephK, extensive parts selection, listening sessions and a LOT of money.
And still you're attributing me such a stupid behaviour...
Maybe you should read with more attention...
I don't get what do you mean here...
None, I've bought all needed parts to made an 'in circuit' test but I still didn't manage to find the time.
At the same time the preliminary tests I've did on the DAC didn't made me more curious but more skeptical instead.
ADA4627, OPA827, OPA140, OPA1641 are all formally better than LM318 but none of them, at least in that application, have the 'magic' of its sound, the 'magic' of the My_Ref sound I love and that I want to keep.
The best opamp and less distant from LM318 seems OPA827 but with serious drawbacks, OPA140 doesn't have such drawbacks but is missing the magic OPA827 has.
At this point only an in circuit test will tell me if the OPA827 drawbacks are mitigated (or present at all) on the My_Ref topology or if the OPA140 (or others) will fare better.
Feedbacks by who tried the OPA827 on the FE like Sontero or Suburra seems very promising in that regard.
In the meanwhile builders can use with confidence the LM318 which sounds great and who is willing to experiment is obviously free to use other opamps leveraging the great work made by JosephK.
Last edited:
Oh, yes! I have missed this in the following post you made 2 years ago. Well, I can only say that your findings were more reliable than I though.
And I confirm it once again. I was left speechless.
But I think you are taking my comments in the wrong light. it's just that I felt you skeptical about the opamp change. That's it.
See, I wasn't imagining things
See Jac reply to me, he thought you did already.
I hope that you will soon, I think your most loyal tweakers are waiting for words for the master.
And please include ADA4627-1A and ADA4627-1B which on paper (read: George measures) looks the best. I found and saved em all.