Hi Rolls
I agree, R-cores are very nice, but not that widely available. Thanks a lot for the link, which I have bookmarked for future use.
Hi Russ and Brian
Sure, I agree that the heat generation with the 15 Volt secondaries is entirely acceptable. I just wanted to be on the safe side, as I'm intending to put everything in a closed case. Your new 9 Volt torroid is an excellent idea, also very useful to power some additional glue logic stuff.
I agree, R-cores are very nice, but not that widely available. Thanks a lot for the link, which I have bookmarked for future use.
Hi Russ and Brian
Sure, I agree that the heat generation with the 15 Volt secondaries is entirely acceptable. I just wanted to be on the safe side, as I'm intending to put everything in a closed case. Your new 9 Volt torroid is an excellent idea, also very useful to power some additional glue logic stuff.
Russ White said:
Russ
I will be pleased to test it thoroughly for you. I have a variety of high quality 176.4 and 192k sources.
Is the LV a SOT23 part; the datasheet is quite unclear about the alternative footprint.
Please post/contact me when you have done the mod and I shall swop and may be buy another one if I can find a use for I2S. But it is spdif to 192k that I need and this will be the strong point of the sorted Twisted Pear Buffalo.
Fred
Re: passive I to V
Hello Mike
I have built my buffalo with all possible outputs. I can use single ended and balanced on IVY, but I can also use the buffalo directly into my Django TVC (transformer volume control).
Now the winner: Direct output to Django, the buffalo works in voltage mode!
What I mean: Don't bother with I/V resistors, do it with a 1:1 transformer, probably from Lundahl. Maybe you won't listen to other i/v stages any more.
Russ says, the distorsion is lower in current mode, but the direct path sounds even better. Gordon Rankin does it the same way!
André
Hello Mike
I have built my buffalo with all possible outputs. I can use single ended and balanced on IVY, but I can also use the buffalo directly into my Django TVC (transformer volume control).
Now the winner: Direct output to Django, the buffalo works in voltage mode!
What I mean: Don't bother with I/V resistors, do it with a 1:1 transformer, probably from Lundahl. Maybe you won't listen to other i/v stages any more.
Russ says, the distorsion is lower in current mode, but the direct path sounds even better. Gordon Rankin does it the same way!
André
Getting electronics to sound as clean as a transformer is quite a challenge - but I believe that it can be done - I hope counterpoint hits that target because in my experience the electronics option can ultimately give better clarity . . . but getting the noise as low as copper wire - that is a target to aim for - 
Re: passive I to V
Hi Mike, I would not use a resistor I/V. As really that will just be a voltage divider.
When used with a resistor this way the DAC is still really in voltage output mode. It is not same sort of device as say a PCM1794A.
You really should not even need it. The voltage output of the ESS chip is just a bit above 2VRMS as it is. Not really any need to lower it.
Cheers!
Russ
mikelm said:
Hi Mike, I would not use a resistor I/V. As really that will just be a voltage divider.
When used with a resistor this way the DAC is still really in voltage output mode. It is not same sort of device as say a PCM1794A.
You really should not even need it. The voltage output of the ESS chip is just a bit above 2VRMS as it is. Not really any need to lower it.
Cheers!
Russ
Re: Re: Re: passive I to V
Well it is a balanced output so both transformer i/p are at same level . . . but this is why I ask andre if he use a ground lift circuit element.
I think this is the classic method - it would just be with a perhaps 10k resistor from primary centre tap to the DAC output ground and then a capacitor across the resistor - perhaps 100uF
I have used this method before but would interested to get some feedback on this.
mike
Hompie Dompie said:
Well it is a balanced output so both transformer i/p are at same level . . . but this is why I ask andre if he use a ground lift circuit element.
I think this is the classic method - it would just be with a perhaps 10k resistor from primary centre tap to the DAC output ground and then a capacitor across the resistor - perhaps 100uF
I have used this method before but would interested to get some feedback on this.
mike
Re: Re: Re: Re: passive I to V
I am glad to see that there are people around with transformers!
A TVC has no ground connection in balanced mode. There are 2 primary windings, but provided just to get +6db output if connected parallel. In balanced mode ground is just routed from the input to the output XLR connector. I leave my ground lift switch in the open position even when listening unbalanced, because this way you have perfect protection against ground loops and every item has, and should have (!) its own earth connected. There is an interesting paper by Jensen.
About S&B TVC's:
My Django has got a Mk2 and even Mk3 is not the last word in resolution. There has just been a privat contest in our country. The contenders were:
S&B Mk3, Silk, Audio Consulting copper and silver.
Interestingly Audio Consulting copper and Silk transformers had about the same quality, S&B was clearly behind and Audio consulting Silver a class of its own.
I think I will go the passive way and scrimp my money to buy a Silver TVC. The person who has bought one does normally not spend a single unnecessary Swiss Franc and he had told me 2 weeks before the test, that he never would buy a Silver TVC regardless of its quality.....
Back to the topic:
I would like to encourage every buffalo user to try a transformer, either a TVC or another very good transformer in front of a "normal" preamp.
I am glad to see that there are people around with transformers!
A TVC has no ground connection in balanced mode. There are 2 primary windings, but provided just to get +6db output if connected parallel. In balanced mode ground is just routed from the input to the output XLR connector. I leave my ground lift switch in the open position even when listening unbalanced, because this way you have perfect protection against ground loops and every item has, and should have (!) its own earth connected. There is an interesting paper by Jensen.
About S&B TVC's:
My Django has got a Mk2 and even Mk3 is not the last word in resolution. There has just been a privat contest in our country. The contenders were:
S&B Mk3, Silk, Audio Consulting copper and silver.
Interestingly Audio Consulting copper and Silk transformers had about the same quality, S&B was clearly behind and Audio consulting Silver a class of its own.
I think I will go the passive way and scrimp my money to buy a Silver TVC. The person who has bought one does normally not spend a single unnecessary Swiss Franc and he had told me 2 weeks before the test, that he never would buy a Silver TVC regardless of its quality.....
Back to the topic:
I would like to encourage every buffalo user to try a transformer, either a TVC or another very good transformer in front of a "normal" preamp.
passiv V/V
I have forgotten to address the dc problem:
There is about 3.15 V dc measured from one output and ground, so there is no dc between the outputs. If you want to use a transformer as I/V (the way I use it is a V/V converter because buffalo works in voltage mode), you need a transformer with 2 primary windings and a ratio of about 1:10. Connect the the primaries in serie and the midpoint to ground, so current is able to flow to ground! On the secondary you will get about (16mA : 10 ) 1.6mA peak current, and if you use a 3.6k Ohm resistor, you will get about 5.7V peak or slightly more than 2Vrms. The Buffalo "sees" (3.6K : 100) around 36 Ohm and is still more in current mode. Because you have a current source, the 3.6k do no harm, they are just parallel.
I have not tried this yet, but it would be interesting to see which is more important.
the buffalo has less distorsion in current mode, but a 1:1 transformer has generally better specs than a 1:10 one.
If you have a TVC, voltage mode seems to be better, because you cannot use a TVC for I/V conversion, you need an additional transformer.
I have forgotten to address the dc problem:
There is about 3.15 V dc measured from one output and ground, so there is no dc between the outputs. If you want to use a transformer as I/V (the way I use it is a V/V converter because buffalo works in voltage mode), you need a transformer with 2 primary windings and a ratio of about 1:10. Connect the the primaries in serie and the midpoint to ground, so current is able to flow to ground! On the secondary you will get about (16mA : 10 ) 1.6mA peak current, and if you use a 3.6k Ohm resistor, you will get about 5.7V peak or slightly more than 2Vrms. The Buffalo "sees" (3.6K : 100) around 36 Ohm and is still more in current mode. Because you have a current source, the 3.6k do no harm, they are just parallel.
I have not tried this yet, but it would be interesting to see which is more important.
the buffalo has less distorsion in current mode, but a 1:1 transformer has generally better specs than a 1:10 one.
If you have a TVC, voltage mode seems to be better, because you cannot use a TVC for I/V conversion, you need an additional transformer.
Russ White said:
Russ
I have now lookes at the eye patterns before and after the comparator.
Everthing is fine up to 96k. At 176.4k, the comparator output is completely broken up. No wonder the Sabre can't latch on.
Looks like the faster LMV7219 is mandatory.
How long will it be for you to intsall the new chip?
Fred
off topic - sorry
After I compared my S&B 102mk 3 to a modified lightspeed type volume control I discarded the transformer. lightspeed was in a different class. Since then I made several improvements to my lightspeed so I would love to compare the silver rock to a well implemented LDR volume control
mike
rolls said:
After I compared my S&B 102mk 3 to a modified lightspeed type volume control I discarded the transformer. lightspeed was in a different class. Since then I made several improvements to my lightspeed so I would love to compare the silver rock to a well implemented LDR volume control
mike
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.