The WM8740 has an internal analog filter with a Fc(-3db) of 195khz.
The Wm8741 also has one but its Fc is 474khz. My guess is they must have decided they did not need to be as aggressive. Probably because they managed to greatly reduce out of band noise, or at least move it up in frequency far beyond audio range. More likely the latter.
Cheers!
Russ
The Wm8741 also has one but its Fc is 474khz. My guess is they must have decided they did not need to be as aggressive. Probably because they managed to greatly reduce out of band noise, or at least move it up in frequency far beyond audio range. More likely the latter.
Cheers!
Russ
Hi Russ
In post #1524 in this thread you mention something I think is a very interesting experiment and one that you thought sounded very good. If I wanted to use this setup as a single ended output, let's say I, on the" left channel IVY" connected and summed the balanced output + signal from the "noninverted side" of the IVY and the - signal from the "inverted side" and use the serial resistors as load balancing resistors? And of course do the same on the right channel IVY. Do you think this would work? And is one LCBPS sufficient to supply two IVY's
Øivind
In post #1524 in this thread you mention something I think is a very interesting experiment and one that you thought sounded very good. If I wanted to use this setup as a single ended output, let's say I, on the" left channel IVY" connected and summed the balanced output + signal from the "noninverted side" of the IVY and the - signal from the "inverted side" and use the serial resistors as load balancing resistors? And of course do the same on the right channel IVY. Do you think this would work? And is one LCBPS sufficient to supply two IVY's
Øivind
xvsop said:
Yes you can do that, if I understand you correctly. 🙂
It would be very similar to the way the Ballsie works. 🙂
Yes one LCBPS can easily power two IVYs. I have one powering 4. I am experimenting with digital XOVER.
Cheers!
Russ
thanks Russ for your quick reply. I should have used schematics, not used to explain stuff like this in english, sorry about that. Keep up the good work!
Cheers
Øivind
Cheers
Øivind
Spartacus said:
I use Hugh Dean's LifeForce monoblocks. They are crazy good. But yes, high on feedback. Originally it has a 1st order filter at 150kHz or something like that. That's definitely audible in the sub 20kHz region. I remember a discussion about this in Greg Ball's thread on diyhifi.org. Graphs and all. But the thread is 60 pages long, so I'm not able to find it.
That filter also means using a ~1nF filter cap, which is way too heavy a load. I spent quite some time finding the optimal value, and ended up around 220pF, ~500kHz. So, no, I'm not going to reduce the Fc. 😉
Russ White said:
Music, sweet music Russ.
😀
Go on do it, do it, do it. Pppppllllllleeeeeeeeeaaaassssssseeeee.
Pretty please.
revintage said:
Yes. As long as there is interest in it, we will keep producing it, and there does seem to be a lot of interest.
Great, will order a dual mono after you are back from vacation.
Also I am keen to try the idea I presented at the audioasylum K&K forum:
Have thought of building a DAC round Wolfson WM8741. My concern is that the DACs should work best with resistor-loading and to get rid of the DC-offset to ground.
When connecting the DAC directly to a transformer, series-resistors are needed. This is not to my liking as my subjective impressing of series-resistors is that dynamics are lost. Also a 1:1 transformer gives us a rather high Zout due to DCR and the series-resistors, although a 2:1 like LL1676 could be used but still Zout will be on the high side due to DCR. Any DC-offset between the +/- outputs might also affect lowend performance.
So a balanced tubestage might be the solution. The DC-offset(relative to ground) is compensated for by the choice of cathode-resistor value. Only low-value resistors are needed to prevent oscillation between the parallelled outputs. Using 6H30 together with a LL1689PPAM 18:1 gives us a gain of just below unity and a sufficiently low Zout.
Off course it could also be built "K&K-style" with two CCSs and balanced parafeed.
Any further thoughts about this idea except for subjective thoughts about the sonics of tubes and transformers?
Also I am keen to try the idea I presented at the audioasylum K&K forum:
Have thought of building a DAC round Wolfson WM8741. My concern is that the DACs should work best with resistor-loading and to get rid of the DC-offset to ground.
When connecting the DAC directly to a transformer, series-resistors are needed. This is not to my liking as my subjective impressing of series-resistors is that dynamics are lost. Also a 1:1 transformer gives us a rather high Zout due to DCR and the series-resistors, although a 2:1 like LL1676 could be used but still Zout will be on the high side due to DCR. Any DC-offset between the +/- outputs might also affect lowend performance.
So a balanced tubestage might be the solution. The DC-offset(relative to ground) is compensated for by the choice of cathode-resistor value. Only low-value resistors are needed to prevent oscillation between the parallelled outputs. Using 6H30 together with a LL1689PPAM 18:1 gives us a gain of just below unity and a sufficiently low Zout.
Off course it could also be built "K&K-style" with two CCSs and balanced parafeed.
Any further thoughts about this idea except for subjective thoughts about the sonics of tubes and transformers?
An externally hosted image should be here but it was not working when we last tested it.
I don't understand why on earth you would put a transformer after the WM8741. 😕 And then a tubestage. 
That sounds awfully complicated for a Vout dac capable of driving a 1K 100pF load.
What's wrong with just a coupling cap?
That sounds awfully complicated for a Vout dac capable of driving a 1K 100pF load.What's wrong with just a coupling cap?
Heja Norge!
Why on earth would you put a capacitor after WM8741
?
Remember the DAC has differental outputs! You must do a conversion from balanced to unbalanced and this will not be possible without either active elements(sand or tubes) and/or transformers.
Unless you go balanced from source to loudspeaker.......... But then you need two caps 😉 !
Why on earth would you put a capacitor after WM8741
? Remember the DAC has differental outputs! You must do a conversion from balanced to unbalanced and this will not be possible without either active elements(sand or tubes) and/or transformers.
Unless you go balanced from source to loudspeaker.......... But then you need two caps 😉 !
revintage said:
That depends on the diff input of the next stage. Some are quite capable of dealing with the offset.
What diff input😉 ?
In my small world diff input is not the usual stuff but off course you are right. The same goes for amps equipped with input caps.
With the "right" diff input that can handle +2,5V DC you could skip the caps all together. But I doubt there are amps that really like it🙂 .
This is another way to use a double-triode and two caps( good PIOs or anything equal) to get an unbalanced out. And YES it has got lots of NFB.
But I think I am in the wrong territory proposing tube circuits
.
In my small world diff input is not the usual stuff but off course you are right. The same goes for amps equipped with input caps.
With the "right" diff input that can handle +2,5V DC you could skip the caps all together. But I doubt there are amps that really like it🙂 .
This is another way to use a double-triode and two caps( good PIOs or anything equal) to get an unbalanced out. And YES it has got lots of NFB.
But I think I am in the wrong territory proposing tube circuits
.Attachments
You are not in the wrong territory but the solution you propose is way too complicated for a DAC that does not need any electronics at the output. You can hook it up without transformer or added electronics.
However, i can understand why you would want this. I made this stage but have no been able to find time to test it.
However, i can understand why you would want this. I made this stage but have no been able to find time to test it.
Attachments
Hi Jean-Paul,
The DAC must be followed by either active elements or transformer if wanting to go unbalanced and using both + and - outputs.
And only transformer, which is what I like, is not an option due to the DACs required series-resistors if your following amp demands feeding from a low impedance source. This is explained in my post above.
The question is: Are the series-resistors necessary together with a transformer as this only loads the DAC at very low frequencies due to its inductance?
About the CF, it was just a joke. I would never dream of using it.
The DAC must be followed by either active elements or transformer if wanting to go unbalanced and using both + and - outputs.
And only transformer, which is what I like, is not an option due to the DACs required series-resistors if your following amp demands feeding from a low impedance source. This is explained in my post above.
The question is: Are the series-resistors necessary together with a transformer as this only loads the DAC at very low frequencies due to its inductance?
About the CF, it was just a joke. I would never dream of using it.
For testing I just followed datasheet instructions of using 1.65 kOhm at the primary windings ( so 2 resistors ). I just added a Zobel secondary again following the datasheet. I am talking about the datasheet of the transformer ! The DAC wll not complain about the transformer loading it I think. The 1.65 kOhm exceed the minimal specifications of 1 kOhm anyway.
You are right about using it unbalanced: it can not be done with both outputs without added electronics. It can be done using only one output but the output voltage is too low then and the DAC is not performing at its best. I wanted it running passive, that's why I made the transformer bal-unbal stage.
You are right about using it unbalanced: it can not be done with both outputs without added electronics. It can be done using only one output but the output voltage is too low then and the DAC is not performing at its best. I wanted it running passive, that's why I made the transformer bal-unbal stage.
revintage said:What diff input😉 ?
This is another way to use a double-triode and two caps( good PIOs or anything equal) to get an unbalanced out. And YES it has got lots of NFB.
But I think I am in the wrong territory proposing tube circuits
I would never discourage an enthusiast from using tubes, but personally I am glad to be free of of that addiction. After my supply of wonderful NOS JAN triodes was burned-up, there was just no satisfaction to be found with newly manufactured tubes. Switching to balanced circuits definitely helped with 'kicking the tube habit'! The wonderfully low noise floor and overall refinement of Russ & Brian's Overture amps was/is very rewarding, and the 8741 inputs directly... 😉
Best,
Frank in Mpls.
Jean-Paul,
Lets do some guesswork going unbalanced with only transformers. I have not checked the data sheets and do not know the data of your trannys so this is not exact:
With a 1:1 transformer Zout = DACZout 2*50ohm + resistors2*1650+ DCR of primary500 + DCR of secondary500= ca 4k
With a 2:1 it then should then be ca 1,5k.
Still to high in my opinion.
Lets do some guesswork going unbalanced with only transformers. I have not checked the data sheets and do not know the data of your trannys so this is not exact:
With a 1:1 transformer Zout = DACZout 2*50ohm + resistors2*1650+ DCR of primary500 + DCR of secondary500= ca 4k
With a 2:1 it then should then be ca 1,5k.
Still to high in my opinion.
Mmm, compared to most tube stages it is not that bad. I think I can live with that. The simplicity of passive stages ( despite this kind of drawbacks ) wins from colouration of added electronics IMO. This is purely subjective ! I now recall I did try the output stage and it sounded good at first glance. It did not fit well in my too small enclosure. I did make another one with Haufe transformers but I really lack time for hobbying so that one has not been touched for some months.
BTW I have chosen Neutrik NTL1. I could not attach the full datasheet as it is too big in filesize. But Google is your friend, it'll find the datasheet.
If you want to try this: I have a spare PCB and only one NTL1 left. Send me a PM in that case. I still need some unobtainable LM4562....
BTW I have chosen Neutrik NTL1. I could not attach the full datasheet as it is too big in filesize. But Google is your friend, it'll find the datasheet.
If you want to try this: I have a spare PCB and only one NTL1 left. Send me a PM in that case. I still need some unobtainable LM4562....