advertisement


Buffalo II - Assembly for the Electronically Bewildered.

Status
Not open for further replies.
So who has upgraded Trident v1 to Trident v3.
I have just swapped a set of v1 for v3 and I am not sure there was any improvement, the latest Tridents were perhaps not as good as the older ones for me. There was very very little in it.
Wondering if they need a burn in?
Swapped the AVCC Trident, it is a Trident? back to a Flea and certainly preferred the Flea.

Tony
 
Messing around the AVCC pins

With a view to changing the regulation on the AVCC rails I decided to do a little messing around and take some measurements.
The results were not what I expected.

During the process I discovered the Placid HD had run very low on shunting current. I had originally set well over 100mA shunt overhead and not adjusted or checked since the first few days of setting up the Buffalo.

This may well explain how I found little change in SQ when comparing Trident v1 to Trident v3, Not enough regulation juice at all. And maybe why Flea performed so well when it was not at all optimised for the position it was in.
The Flea is separately supplied and its addition gave the Placid a comfortable 160mA shunt overhead.

Somehow the Placid went adrift / got altered...
Placid now has 575mA CC and 110mA shunt


Back to the messing around AVCC

The TP AVCC module draws about 160mA in total from the Vd rail.
I measured it.

From some were on the internet, each converter is 4mA.
So four converters per channel is 16mA per channel, 32mA in total.

AVCC current per channel is about 40mA
I measured it, both muted and playing music it seems a steady current on the multimeter.

Each converter is actually a pair of differential current converters so the current should be constant at all times... Am I right here?

Actually the AVCC current on the left converter was 43mA and the right converter 47mA

So a few things get whirling in my mind.

The TP AVCC module has quite A high standing shunt current at about 50%.

Why does each converter channel pull more than twice the expected current?

Why are left and right channels pulling differing currents? I expected them to be exactly the same.

Any thoughts gents
Tony
 
On the subject of Placid "drift"; I found mine would alter a fair bit, requiring adjustment several times before finally settling down to a degree of stability. I'm glad I replaced them with the PH regs, which seem pretty bulletproof, and his Trident/AVCC replacements made a very worthwhile improvement to sound quality.
 
Tony,
I finally pulled the trigger on three (large :) ) custom cases from Lincoln Bins that you put me in touch with.
The wallet will take a while to recover - but it will be worthwhile.

Then comes the real fun part of making all those psu's talk to the dac boards. It's been that long that I need to completely re read all of the instructions and this thread!
I also have the Wave Io board on way so that is yet another PH ps needed..., and another flippin box.
 
For my part, I've discovered the importance of shielding. The Buff II has locked with some sources but suffered from unstable locks ("stuttering") with others. Nothing made any sense. The latest misadventure has been with a new AppleTV. Audio was unlistenable even though the video was gorgeous.

The solution?

The AppleTV is now wrapped in Al foil with just the front showing. :rolleyes:
 
Messing around the AVCC pins

With a view to changing the regulation on the AVCC rails I decided to do a little messing around and take some measurements.
The results were not what I expected.

During the process I discovered the Placid HD had run very low on shunting current. I had originally set well over 100mA shunt overhead and not adjusted or checked since the first few days of setting up the Buffalo.

This may well explain how I found little change in SQ when comparing Trident v1 to Trident v3, Not enough regulation juice at all. And maybe why Flea performed so well when it was not at all optimised for the position it was in.
The Flea is separately supplied and its addition gave the Placid a comfortable 160mA shunt overhead.

Somehow the Placid went adrift / got altered...
Placid now has 575mA CC and 110mA shunt


Back to the messing around AVCC

The TP AVCC module draws about 160mA in total from the Vd rail.
I measured it.

From some were on the internet, each converter is 4mA.
So four converters per channel is 16mA per channel, 32mA in total.

AVCC current per channel is about 40mA
I measured it, both muted and playing music it seems a steady current on the multimeter.

Each converter is actually a pair of differential current converters so the current should be constant at all times... Am I right here?

Actually the AVCC current on the left converter was 43mA and the right converter 47mA

So a few things get whirling in my mind.

The TP AVCC module has quite A high standing shunt current at about 50%.

Why does each converter channel pull more than twice the expected current?

Why are left and right channels pulling differing currents? I expected them to be exactly the same.

Any thoughts gents
Tony

Tony

This echoes my experience with the latest Tridents. When I first installed them in my Buffalo2 I reported disappointment with them. My view has changed slightly, but it has taken some radical surgery. Previously I had all mains PSU stuff in a separate case, with a longish (too long really) multipin cable connecting to the DAC and IV stages. I found that by moving the whole PSU and DAC into one case I was able to minimise sag on the 5.25 line feedng the Buffalo. I did have to up the shunt current on my Placid (old version) to pretty much it's max (630ma, iirc, by measurement) but that includes feeding not only the Tridents and AVCC regs on the Buffalo but also a separate Trident V1 3.3v for my I2S galvanic isolation stage.

Now the Tridents are sounding better, but I am hoping to move away from the TPA power supply/reg approach soon for something a little more elevating hopefully.

I did not measure current one the AVCC module but did not that under voltage sag on the supply side, the LED on one half of the AVCC supply would become much dimmer than the other half. Sounds similar to what you've observed.

My Placid supply for the DAC runs pretty warm now, so I need to move to a higher current high performance reg for that pretty soon.

Overall though it does sound quite compelling now and I am hoping there's a little more to come.

Mark
 
On the subject of Placid "drift"; I found mine would alter a fair bit, requiring adjustment several times before finally settling down to a degree of stability. I'm glad I replaced them with the PH regs, which seem pretty bulletproof, and his Trident/AVCC replacements made a very worthwhile improvement to sound quality.

Hi Tony, so mine just probably drifted, began to think I was going mad.

Tempted to try the PH regs as you and others recommend them but are also tempted to dig my own pit with some Salas and Fleas and maybe a ALWSR and bite a learning curve again. I have lots of tricks to try with the Fleas.

I also considered that my Arduino could easily monitor the Shunt reg currents...
 
Tony,
I finally pulled the trigger on three (large :) ) custom cases from Lincoln Bins that you put me in touch with.
The wallet will take a while to recover - but it will be worthwhile.

Then comes the real fun part of making all those psu's talk to the dac boards. It's been that long that I need to completely re read all of the instructions and this thread!
I also have the Wave Io board on way so that is yet another PH ps needed..., and another flippin box.

Be nice and interesting for me if you could summarise what your doing in one post, I was on sabbatical at the time of your original posts.

Tony
 
For my part, I've discovered the importance of shielding. The Buff II has locked with some sources but suffered from unstable locks ("stuttering") with others. Nothing made any sense. The latest misadventure has been with a new AppleTV. Audio was unlistenable even though the video was gorgeous.

The solution?

The AppleTV is now wrapped in Al foil with just the front showing. :rolleyes:

Hi flash, pleased your sorted.
Did you experiment with the phase lock settings and notch? (DPLL)
Closing the lock loop to tightly on hi sample rates can give trouble as can a cold Buffalo, the crystal drifts around a little until warm.
You got pictures of your screening?

Tony
 
Tony

This echoes my experience with the latest Tridents. When I first installed them in my Buffalo2 I reported disappointment with them. My view has changed slightly, but it has taken some radical surgery. Previously I had all mains PSU stuff in a separate case, with a longish (too long really) multipin cable connecting to the DAC and IV stages. I found that by moving the whole PSU and DAC into one case I was able to minimise sag on the 5.25 line feedng the Buffalo. I did have to up the shunt current on my Placid (old version) to pretty much it's max (630ma, iirc, by measurement) but that includes feeding not only the Tridents and AVCC regs on the Buffalo but also a separate Trident V1 3.3v for my I2S galvanic isolation stage.

Now the Tridents are sounding better, but I am hoping to move away from the TPA power supply/reg approach soon for something a little more elevating hopefully.

I did not measure current one the AVCC module but did not that under voltage sag on the supply side, the LED on one half of the AVCC supply would become much dimmer than the other half. Sounds similar to what you've observed.

My Placid supply for the DAC runs pretty warm now, so I need to move to a higher current high performance reg for that pretty soon.

Overall though it does sound quite compelling now and I am hoping there's a little more to come.

Mark

Hi Mark, interesting similarity, actually my whole SQ had drifted off along with the Placid, but it was gradual and hard to notice. just I thought the initial impact of the Buffalo had just worn off. Its back on song again now though.

I will struggle to get all my kit in one box and have thought to put the transformers feeding HackerCAPs in one box and regulators and Buffalo in the other box. I do not want the transformer secondary rectifier pulses traveling from box to box, and I want the Regulators right by the Buffalo. Adding the HackerCAPs neatly solves the problems and may improve on the shunt regulator performance also.

I am sure there is more to come, in fact Suffolk Tony has already got it :D

Tony
 
Still not making sense of my AVCC currents perhaps I will have a 'moment' around the garden today :confused:

Oh I forgot to mention that my Buffalo-III is now directly coupled to the HackerNAPs :eek::

That is directly as in DC coupled :eek::eek::eek:

Not sure how long I can live with the suspense though an may be adding a coupling cap :(
 
Mines been directly coupled for a while as well. I do check it every couple of weeks though! It does drift as bit. I'm thinking of installing one of those old fashioned bias meters like valve amps sometimes have :). It does sound good without caps!
 
I think locating the DC regulation (but not necessarily the rectification) close to the DAC and output stages is to be preferrred. I certainly had more than 0.5v sag with connectors and cables between two boxes (though I admit my cable was unnecessarily long at 2 metres) from the output junction on the Placid and the input at VD on the Buffalo board.

Placid voltage also wanders (downward by about 0.2v) from cold to toasty.

ALWSRs would probably bring benefits in certain locations but Martin Clark and I have been discussing alternative approaches - the only reason I haven't followed Suffolk Tony to Paul Hynes already!

If you can get away with a dc-coupled output I think you can gain significant sonics. Not viable in my system even with the offset trimmed on the Legato.
 
ALWSRs would probably bring benefits in certain locations but Martin Clark and I have been discussing alternative approaches - the only reason I haven't followed Suffolk Tony to Paul Hynes already!

Pray do tell more :D

How come you have a direct line to god?

I consider loosely AWSLR for the Legato and also STR, STR keeps creeping back despite its shortcomings.

Is Martin about to come up with a new regulator? I am sure its at the back of his mind.


I think locating the DC regulation (but not necessarily the rectification) close to the DAC and output stages is to be preferrred. I certainly had more than 0.5v sag with connectors and cables between two boxes (though I admit my cable was unnecessarily long at 2 metres) from the output junction on the Placid and the input at VD on the Buffalo board.

Yes all on box is best but if we use two boxs we need local regulators right by the Buffalo in one box and transformers / rectification / smoothing in the other.

I certainly had more than 0.5v sag with connectors and cables between two boxes (though I admit my cable was unnecessarily long at 2 metres) from the output junction on the Placid and the input at VD on the Buffalo board.

Connectors have minimal resistance so bigger cables?
But 0.5 volt over 500mA implies there is about a whole ohm kicking around in your interconnects, 500mOhm per meter

Even piddly cat5 cable is about 0.1 Ohm per meter?
 
Tony

My interconnecting cable was 10 metres (terminated in D-type 15 pin, I was running separate feeds for each section of the DAC & I/V and SPDIF receiver and and and) at one point from a separate regulated supply), completely unnecessary but the shortest I could source at the time. I was surprised to discover the drop off was so high, so I tried an alternative SPMS good for 5 or more amps and stil saw the same drop off. Binned the separate box approach at that point.

I'm not exactly sure what Martin has in mind, but we've been talking about doing this for a little while. Really down to Martin finding some head time to put his thoughts on paper at this stage. All in good time...

I can't recall what current the ALWSRs can sustain, but I'd thought of using one set to about 5.25v to replace the Placid supplying the Buffalo. That might make life easier for the Tridents. The Placid is rather on the edge in my set up and the smallish reservoir cap between the rectrifier and the regulation circuit is too small to provide meaningful support, so I am probably losing some goodness right there. When I get five minutes I may do a little surgery and stick 10000uf in there instead of the 1000uf currently installed. This is one of the improvements with the Placid HD I believe.

Mark
 
The ALWSR manual claims "As an absolute minimum the regulator can deliver 200mA, most units exceed this specification by a significant amount (500mA+)." However, in my experience, I've found it to be a bit feeble. Given local regs at the Buffalo in the form of Tridents or PH regs, my strategy is to go for more power in the pre-reg, even at the expense of some noise.
 
A simple upgrade on stock TP gear seems to be removing the Placid rectifier pre filter and replacing with the smaller of the HackerCAP boards. Two HackerCAPS would feed the Placids for a whole Dual mono. Downside is only the amount of real estate taken by the HackerCAPs.
 
Thanks to Suffolk Tony :D:D:D


036.jpg
 
Status
Not open for further replies.


advertisement


Back
Top