Hi Tom,
4.5 uA is actually quite high and wouldn't allow for removing C9, at least as I understand it. I was looking at the Sparkos features and see that it is listed as "100% Class A biasing". That may explain both the high bias point and why it sounds good in other applications. For reference, the LM318 has a max bias current of 250 nA and the ADA4627, OPA827, and OPA828 are in the 1 to 10 pA range.
Seems like a really nice opamp for other applications. I'm glad people are still trying things.
Jac
4.5 uA is actually quite high and wouldn't allow for removing C9, at least as I understand it. I was looking at the Sparkos features and see that it is listed as "100% Class A biasing". That may explain both the high bias point and why it sounds good in other applications. For reference, the LM318 has a max bias current of 250 nA and the ADA4627, OPA827, and OPA828 are in the 1 to 10 pA range.
Seems like a really nice opamp for other applications. I'm glad people are still trying things.
Jac
Last edited:
C1 condensatore di ingresso
Buonasera Dario,
è possibile montare un condensatore di ingresso da 0,1 microf ?
uso il MyRef fe per pilotare largabanda con risposta in frequenza da 130hz a 18000hz
please post an english translation. This was done using google translate.
Buonasera Dario,
è possibile montare un condensatore di ingresso da 0,1 microf ?
uso il MyRef fe per pilotare largabanda con risposta in frequenza da 130hz a 18000hz
please post an english translation. This was done using google translate.
Last edited by a moderator:
Hi guys
I need your help.
Until now, I used an android box with a usb dac directly connected to the myref ampli. Everything ok.
I have 2 trafo and the 2 channels are tottaly floating. I didn't connect PGND to nothing and the same for HSGND.
Now I build a volume control circuit and put it in the middle between the DAC and the myref.
The volume control has one circuit for both channel with one trafo (and one GND). ALso in this case the GND is not connected to anything.
All tests that I did on the preampli (isolated) are ok. Low distortion and clean output.
But when I connect the preampli to myref and play music, I can hear distortion. No hum, but strong distortion.
I did a lot of test and finally I connected the DAC to the USB of my PC, DAC to preampli and preampli to myref.
In this situation I can still hear distortion, but doing this test I saw that if I connect the GDN of volume control to the hearth the distorsion disappear... and there is no hum.
Even if I was surprised, I had tought that I solved the problem. So I connected the dac to the android box and restart everything.... HUM! Of course there is hum...
I tryed to power the android box with a power bank, but hum is still there.
So my understanding is that the problem is related to 3 floating GND (1 volume control, 1 left channel of myref, 1 right channel of myref).
Am I right?
If yes, how can solve? I tought to connect PGND of both myref channel to the GND of volume control. Is this correct?
I need your help.
Until now, I used an android box with a usb dac directly connected to the myref ampli. Everything ok.
I have 2 trafo and the 2 channels are tottaly floating. I didn't connect PGND to nothing and the same for HSGND.
Now I build a volume control circuit and put it in the middle between the DAC and the myref.
The volume control has one circuit for both channel with one trafo (and one GND). ALso in this case the GND is not connected to anything.
All tests that I did on the preampli (isolated) are ok. Low distortion and clean output.
But when I connect the preampli to myref and play music, I can hear distortion. No hum, but strong distortion.
I did a lot of test and finally I connected the DAC to the USB of my PC, DAC to preampli and preampli to myref.
In this situation I can still hear distortion, but doing this test I saw that if I connect the GDN of volume control to the hearth the distorsion disappear... and there is no hum.
Even if I was surprised, I had tought that I solved the problem. So I connected the dac to the android box and restart everything.... HUM! Of course there is hum...
I tryed to power the android box with a power bank, but hum is still there.
So my understanding is that the problem is related to 3 floating GND (1 volume control, 1 left channel of myref, 1 right channel of myref).
Am I right?
If yes, how can solve? I tought to connect PGND of both myref channel to the GND of volume control. Is this correct?
Ok, I found the problem.
When I connect the DAC to myref, the cold pin is connected to 0V.
My preamp is based on differential schema, so when I connect the DAC to preampli no pin are connected to GND.
If I connect the cold pin of the DAC to preampli GND the distortion disappears.
Now my idea is to add a signal GND connector on the back side of my preampli and then connect the external part of 2 RCA to it.
When I connect the DAC to myref, the cold pin is connected to 0V.
My preamp is based on differential schema, so when I connect the DAC to preampli no pin are connected to GND.
If I connect the cold pin of the DAC to preampli GND the distortion disappears.
Now my idea is to add a signal GND connector on the back side of my preampli and then connect the external part of 2 RCA to it.
Hello George (other knowledgeable members can answer as well). I hope you and your family are well during this second wave the world is suffering with.
I’ve always been a bit perplexed at the grounding of the heatsinks (being at the same potential as the chassis and PGND). The LM3886 package is insulated and it’s the only thing touching the heatsink (in the simplest build). Why must it be grounded (at the same potential as the chassis)?
Thank you,
Pete
Pete,
In the Myref topology, the input pins of the LM3886 see a relatively high impedance (remember, the feedback resistances are 22k; 47k)
The chip sticks out of the circuit. So no schielding from the ground plane below. The metal tab does give some, but it is not ground, it's - Vee. (there is a tab even in the TF package) Which is signal dependent, even. Depends on the bypassing efficiency.
Then, it is fixed by the screw on a big piece of metal block.(Heatsink) If that metal block is not directly and effectively shortened to signal ground, then it's a big ugly antenna..
So this all just helps in conducting unwanted uncontrolled feedback into the input pins; radio pickup from the environment..
The chip is still active into the low MHz range..
So if You add all this up: deaden that chunk of metal as good as You can..
In the Myref topology, the input pins of the LM3886 see a relatively high impedance (remember, the feedback resistances are 22k; 47k)
The chip sticks out of the circuit. So no schielding from the ground plane below. The metal tab does give some, but it is not ground, it's - Vee. (there is a tab even in the TF package) Which is signal dependent, even. Depends on the bypassing efficiency.
Then, it is fixed by the screw on a big piece of metal block.(Heatsink) If that metal block is not directly and effectively shortened to signal ground, then it's a big ugly antenna..
So this all just helps in conducting unwanted uncontrolled feedback into the input pins; radio pickup from the environment..
The chip is still active into the low MHz range..
So if You add all this up: deaden that chunk of metal as good as You can..
Last edited:
It's C13 and R13. C13 should be at least 1uF unless you want to use that as a first order hi-pass filter, which is what I do with my amps because they power only the midranges and tweeter. Making C13 much larger than 1uF does not gain much, if anything.
C13 is by far the most important part for determining the quality of sound. Do NOT cut costs there.
Peace,
Tom E
C13 is by far the most important part for determining the quality of sound. Do NOT cut costs there.
Peace,
Tom E
Merry Christmas everyone, like other HiFi Collective customers I received a coupon that I will not use (unfortunately I have already made my purchases ...).
10% discount. Here is the message and the voucher code:
"Hi guys, I just wanted to personally take this opportunity to wish you a Merry Christmas and a Happy New Year on behalf of the HiFi Collective team.
The 10% off Sale starts NOW and finishes at Midnight (00:00), 31st December 2020.
The voucher code is
BOXDAY20
Remember to click ADD COUPON "
If you need to buy Mundorf M-Lytic, M-Resist, Amtrans, Charcroft Z-foil, etc. in europe, HiFi Collective can be a good supplier.
A warm greeting
Ciao
Giacinto
10% discount. Here is the message and the voucher code:
"Hi guys, I just wanted to personally take this opportunity to wish you a Merry Christmas and a Happy New Year on behalf of the HiFi Collective team.
The 10% off Sale starts NOW and finishes at Midnight (00:00), 31st December 2020.
The voucher code is
BOXDAY20
Remember to click ADD COUPON "
If you need to buy Mundorf M-Lytic, M-Resist, Amtrans, Charcroft Z-foil, etc. in europe, HiFi Collective can be a good supplier.
A warm greeting
Ciao
Giacinto
I've only started to order parts for my FE. For C13, I ordered Mundorf M-CAP® Supreme Capacitors 2% 600 VDC Capacitors (this was part of a larger order this past September). I see now that the preferred part is listed as 250 VDC. Is this likely to make any difference in SQ ?
Thanks.
The voltage capacity of a film capacitor should not make any difference in the sound, unless it does. There are too many factors in what makes a certain part in a certain circuit sound a certain way to provide blanket generalizations.
There are few methods of measuring such differences other than by simply listening carefully. Typically, the voltage rating of a film capacitor is a function of the thickness of the film dielectric between the layers of conductive material: a higher voltage rating implies thicker insulation. In a film cap, that could make a small contribution to the sound by affecting the dissipation factor of the insulation. One would expect thinner insulation to have a lower DF, which should improve the sound. Except some subjective reviewers indicate that, all else being identical, higher voltage film caps sound better than their lower voltage counterparts. So there is a contradiction, and I can't say whether I believe it's true or whether the testing is faulty or biased, or, as sometimes happens, the science does not explain the difference. Whatever the truth is, the difference is sure to be minute either way.
The only things for certain are that a higher voltage cap will handle higher voltages (duh) before failing, it will be physically larger, and it will probably cost more than an identical brand and model with a lower voltage capacity.
In the C13 implementation, the cap usually sees less than one volt and almost never more than two, so voltage rating is meaningless for functionality. Use the smallest cap that gives the best sound. Capacitance value, however, is important for low frequency integrity. Less than 1uF will begin to impair lows.
Peace,
Tom E
There are few methods of measuring such differences other than by simply listening carefully. Typically, the voltage rating of a film capacitor is a function of the thickness of the film dielectric between the layers of conductive material: a higher voltage rating implies thicker insulation. In a film cap, that could make a small contribution to the sound by affecting the dissipation factor of the insulation. One would expect thinner insulation to have a lower DF, which should improve the sound. Except some subjective reviewers indicate that, all else being identical, higher voltage film caps sound better than their lower voltage counterparts. So there is a contradiction, and I can't say whether I believe it's true or whether the testing is faulty or biased, or, as sometimes happens, the science does not explain the difference. Whatever the truth is, the difference is sure to be minute either way.
The only things for certain are that a higher voltage cap will handle higher voltages (duh) before failing, it will be physically larger, and it will probably cost more than an identical brand and model with a lower voltage capacity.
In the C13 implementation, the cap usually sees less than one volt and almost never more than two, so voltage rating is meaningless for functionality. Use the smallest cap that gives the best sound. Capacitance value, however, is important for low frequency integrity. Less than 1uF will begin to impair lows.
Peace,
Tom E
truepaul...the BOM for FE board version 1.72 (the one I have) called for the 600V version of the Mundorf Supreme "Classic" . It doesn’t come in a smaller voltage as far as I can see.
I also used the Mundorf 1.0uF 1000Vdc MCap® Supreme SilverOil in another build and it sounds wonderful as well.
As madisonears just mentioned, physical size has an impact on choosing C13 as well.
Regards, Pete
Last edited:
Must be a Yankees fan madisonears.....hahahahahaha....Yogi-isms are profound!
Pete
A question for those of you who have played around with the Isabellenhutte resistor in R3.
A couple of months ago, I put the Isa's into my main amp (v1.72 board) with some lead adaptation and a brass plate for damping at the suggestion of George. I didn't get a direct comparison to my Quartets because I was diagnosing a problem, but the Isa's sound very good. I haven't tried the Mundorf's yet, but I am looking forward to trying them on the new board.
I am noticing some temperature related behavior that may be related to the Isa's and wondered if anyone else has noticed the same? When I first turn on the amp (cold start) it sounds a little bit harsh and, possibly, a bit louder. I don't want to overstate this, the effect is quite small. But after about 10 minutes, the sound is richer and smoother. For example, on the higher frequency notes of a piano, when cold there is a bit of an edge to the notes, as if harmonics were a little emphasized or a touch of distortion. After warming up, the harmonics are both smooth and warm.
I am only guessing that this could be the Isa's. I did have a problem with my balanced line receiver power supply which I fixed and it could be involved.
I am also curious if the Mundorf's show something similar. After all, they are quite similar in construction and use the same resistance material.
While we are talking about R3, there was some early discussion about non-inductive wirewound resistors for R3, but no following comments. Can we assume that they didn't turn out to be as good a sounding choice as the Mundorf's?
Jac
A couple of months ago, I put the Isa's into my main amp (v1.72 board) with some lead adaptation and a brass plate for damping at the suggestion of George. I didn't get a direct comparison to my Quartets because I was diagnosing a problem, but the Isa's sound very good. I haven't tried the Mundorf's yet, but I am looking forward to trying them on the new board.
I am noticing some temperature related behavior that may be related to the Isa's and wondered if anyone else has noticed the same? When I first turn on the amp (cold start) it sounds a little bit harsh and, possibly, a bit louder. I don't want to overstate this, the effect is quite small. But after about 10 minutes, the sound is richer and smoother. For example, on the higher frequency notes of a piano, when cold there is a bit of an edge to the notes, as if harmonics were a little emphasized or a touch of distortion. After warming up, the harmonics are both smooth and warm.
I am only guessing that this could be the Isa's. I did have a problem with my balanced line receiver power supply which I fixed and it could be involved.
I am also curious if the Mundorf's show something similar. After all, they are quite similar in construction and use the same resistance material.
While we are talking about R3, there was some early discussion about non-inductive wirewound resistors for R3, but no following comments. Can we assume that they didn't turn out to be as good a sounding choice as the Mundorf's?
Jac