• Disclaimer: This Vendor's Forum is a paid-for commercial area. Unlike the rest of diyAudio, the Vendor has complete control of what may or may not be posted in this forum. If you wish to discuss technical matters outside the bounds of what is permitted by the Vendor, please use the non-commercial areas of diyAudio to do so.

Twisted Pear Audio - Buffalo32S (ES9018 DAC)

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Its been running in for about 24hours now.

Its like what the original Sabre gave compared to other dacs but much more so :D

This one is very smooth indeed! more texture than the old Buffalo.
Theres lots of detail I didn't notice before with some music, some instruments are maybe a little thick with soft edges but I think the dac still needs some more days running in? I also notice theres a lot of Oscon SVP caps including on the analogue decoupling which probably take longer to burn in?

Other thing which I find different is the depth from front to back is much deeper, theres lots of info to be heard and it digs right down into the recordings
 
mikelm said:


Some respected professionals - like dennis morecroft of DNM think that a wooden or plastic enclosure has inherent advantages over a metal casing especially a magnetic metal casing. I would be interested to hear if your DAC sounds different if you mount it in a metal box

cheers

mike

Only downside with wood or plastic is the lack of shielding from RF, maybe worth lining the inside with foil or something similar
 
mmm - then there was the case of peter from south africa who's chip amp sounded great in it's 1/2" ali box until he put the lid on - then he said the sound audibly deteriorated.

So I think avoiding circuits for induced currents may be worth checking out.

p.s. my buffalo is already in a foil lined plastic & MDF box ( no lid mostly ) but I never made the comparison with a regular metal box - hence my post :)
 
mikelm said:
mmm - then there was the case of peter from south africa who's chip amp sounded great in it's 1/2" ali box until he put the lid on - then he said the sound audibly deteriorated.

So I think avoiding circuits for induced currents may be worth checking out.

p.s. my buffalo is already in a foil lined plastic & MDF box ( no lid mostly ) but I never made the comparison with a regular metal box - hence my post :)

He may have heard an improvement especially if the transformer was in the same case, beauty with this hobby is that your free to try these things and compare then post your results :)

Only thing to consider is that something like a dac maybe more sensitive to RF than say a chip amp
Also I've heard that a bit of stray RF had the effect of added air to the high frequencies ;)
 
leo said:
Its been running in for about 24hours now.

Its like what the original Sabre gave compared to other dacs but much more so :D

This one is very smooth indeed! more texture than the old Buffalo.
Theres lots of detail I didn't notice before with some music, some instruments are maybe a little thick with soft edges but I think the dac still needs some more days running in? I also notice theres a lot of Oscon SVP caps including on the analogue decoupling which probably take longer to burn in?

Other thing which I find different is the depth from front to back is much deeper, theres lots of info to be heard and it digs right down into the recordings

Hi Leo, I've been running mine for about half that time and although it is much too early to pass any serious comments on sound, I would say my initial reactions were the same as yours and would also add that there is perhaps something else not quite right to my ears...

It is hard to put my finger on, maybe it has to do with the slight softness or thickness or that it just seems a little forced. Like it is trying to hard or doesn't have enough power? OK so I maybe making a pigs ear of describing this, sorry... it is not easy. It is perhaps like the sound of solid core silver cables before they are run in e.g. a little uptight?

I using 48/24 from a Mac Mini (it won't go any higher without losing lock - but it will on my iMac though), so that maybe part of it. I do have a bigger UK transformer of 30va I could try, rather than the US 15va I'm currently using, but I'm not sure if this is going to make much difference for the IVY sections?

Anyway, I'm of course going to give it a lot longer before any more comments, just thought people might like to know. The old Buffalo seemed to take quite a bit of time to come on song... and when it did, it may have lacked some of the detail, depth and clarity in complex passages of the new DAC, but it was also more relaxed than the new one currently is, here's hoping ;-)

Russ - back to some Miles...

(MacMini via optical, TOSLINK/LCDPS/LCBPS, Volumite on the DAC, direct into Welborne Terraplane 300b's monoblocks, into a pair Beauhorn B2.3's.)
 
rjbaldwin said:


Hi Leo, I've been running mine for about half that time and although it is much too early to pass any serious comments on sound, I would say my initial reactions were the same as yours and would also add that there is perhaps something else not quite right to my ears...

It is hard to put my finger on, maybe it has to do with the slight softness or thickness or that it just seems a little forced. Like it is trying to hard or doesn't have enough power?


I know what you mean there mate, I'm wondering if it could be the SVP caps on the analogue decoupling?
I've had a long listen today on my own and have been hearing a slight colouration/distortion which I think maybe the same as the forced sound your getting? its not a high frequency harshness, more in the midrange, upper midrange
Cymbals sound much further back compared to my modified first Buffalo

I'll also give mine more time running in, from past experience with things like BG caps the sound of those go up and down whilst breaking in, maybe these SVP's are similar? from what I've read they are Polymer based, I don't have much experience with those
 
leo said:
...I've had a long listen today on my own and have been hearing a slight colouration/distortion which I think maybe the same as the forced sound your getting? its not a high frequency harshness, more in the midrange, upper midrange
Cymbals sound much further back compared to my modified first Buffalo...

Yes Leo, that could well be it, the same as I am experiencing.
 
rjbaldwin said:


Yes Leo, that could well be it, the same as I am experiencing.


Hey guys,

I worked with Leo yesterday to identify what was going with his setup.

What it boils down to is this. There are two links under the BUF32S board that link the two AGNDs to DGND.

The reason the links were not populated was because we wanted people to have some freedom in how they approached joining the GNDs. We also had some differing advice and wanted to verify a couple of different approaches.

In hindsight we probably would have just put jumpers on there.

Bottom line is (and all of this is in the manual which will be out shortly) all GNDs need to be joined one way or the other, at the L1 and L2 spot under the board works great (just a short wire soldered across the pads or even just a solder blob).

You could also join the GND with wires at the terminal blocks or even at the power supplies. Any of those approaches will work, but you need to be sure to join them one way or the other. At L1 and L2 seems to be the best solution so far.

When Leo did this (I hope he does not mind me saying so) he found the sound to be much better and his muting issues went away. Both of these make perfect sense.

So to wrap up, at least for this initial run of boards the best thing to do is the join the GND signals on the board by shorting the pads of L1 and L2.

I wanted to get this out here quickly since I know many of you are already using the DAC.

Cheers!
Russ
 
BUF32S output options

One thing you will notice on the BUF32S is that there are some parallel lines at some of the traces on the board.

These lines mark spots where you can cut traces for some optional features.

I have attached an image (this information is in the manual too) that indicates the location of the marks.

The marks circled in red when cut enable 21R output resistors that are normally shorted. Cutting those traces makes the 21R effective. If you are going into an amplifier I would definitely suggest cutting at the locations marked. This is especially true for capacitive cables/loads. You will hear a significant improvement.

If you are going directly out to headphones then you will want to leave the trace uncut. The BUF32S very easily drive almost any headphone. Even 32ohm cans.

The positions marked blue "X" show spots where you can cut traces to allow you to supply your own reference voltage for AVCC. If you cut at the marks specified you can supply your own reference voltage for AVCC. Do not do this unless you absolutely know what you are doing. The standard reference is excellent and heavily filtered.

The reason I included the option was that you can actually slightly improve DNR/THD by supplying a reference higher than 3.3V. :) But you need to be very very careful!!! Do not attempt it unless you have read the DS and understand the risks.

Cheers!
Russ
 

Attachments

  • silk_only_output_traces_cuts_1000.jpg
    silk_only_output_traces_cuts_1000.jpg
    95.4 KB · Views: 1,017
Re: BUF32S output options

Russ White said:
... So to wrap up, at least for this initial run of boards the best thing to do is the join the GND signals on the board by shorting the pads of L1 and L2...
Just to be absolutely clear, join the two solder points, joined by two white lines (the points are very close together) at L1 and L2?

Russ White said:

...The marks circled in red when cut enable 21R output resistors that are normally shorted. Cutting those traces makes the 21R effective. If you are going into an amplifier I would definitely suggest cutting at the locations marked. This is especially true for capacitive cables/loads. You will hear a significant improvement...

Russ, I'm running directly into some 300B valve monoblock power amps, single ended, from the Buffalo (I just turn the DAC on first at the moment - no thumb-busters set up yet - awaiting manual). Would it be beneficial for me to cut the traces, I presume cut between the lines, at the 4 (red) places you've indicated? Can I do this from the top of the board?

Many thanks, I thought the sound wasn't quite right...
 
I thought it best to leave it and let you update:)
Anyway thanks to you lock etc is now spot on!

It certainly sounds better than it did in my set up at least with the grounds now linked

I was looking on the scope last night to see if I could see any problems (this was before trying the AG and DG mod)

Few things I noticed

The first is that 20k sinwaves look better on the Buffalo32s, at 20k on the old Buffalo it looked odd, like one of those spirograph drawings, with the new Buffalo its a solid sine wave.
I don't have any higher than 20k or I would have seen how high it goes before it starts to roll off

The other thing I noticed was that 1k square now has pre and post ringing on the newer Buffalo, I presume this is down to the OS in the ES9018?

Funnily enough the grounding made no difference to the quick measurements
 
Hot Chips

IC 10 and 11 are running very warm on my B-32 board. I can't hold my fingers on the chips for more than 30 seconds. This is without anything connected to the analog out and the SPDIF connected but the CD player is on pause. They are hot enough that I'd want to put a heat sink on them. Is this normal?

I haven't gotten any audio out of the board, I haven't had a chance go back through the past posts to find the switch positions. Right now they are as shipped, sw 1-4 in the center position. SPDIF set to on.

RossG
 
Re: Re: BUF32S output options

rjbaldwin said:

1) Just to be absolutely clear, join the two solder points, joined by two white lines (the points are very close together) at L1 and L2?



2) Russ, I'm running directly into some 300B valve monoblock power amps, single ended, from the Buffalo (I just turn the DAC on first at the moment - no thumb-busters set up yet - awaiting manual). Would it be beneficial for me to cut the traces, I presume cut between the lines, at the 4 (red) places you've indicated? Can I do this from the top of the board?

Many thanks, I thought the sound wasn't quite right...

1) Yes. The pads are close enough that its fairly easy to to create a solder bridge if you like.

2) Yes I would cut those 4 traces. Those only place to cut them is at the top of the board were the parallel white lines are on the silk. The OPA1632 does not like a capacitive load in the least. The 21R effectively isolates it from the capacitive load (cables).
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.