Tír Na HiFi

abraxalito wrote: ↑Mon Feb 20, 2017 1:48 am
The transformer is really up to you for the ratio you need. It needs to match the maximum output of the DAC to the maximum input of the amp. I'm assuming you're still using the TVC on the output of the amp.

atm lowered 3875 gain to x11, removed ac coupling cap before trafo and using trafo as autoformer

remember someone recommending Polystyrene for low pass filter

Be careful with polystyrene. Attached a picture from PARTS CONNEXION.

Those little plastic things that are everywhere are bad. Allen Wright had a diatribe against them in his PREAMP COOKBOOK.

The Russian teflons are cheap and sound good. Just took a look and do not see them anywhere.

RELCAP has RT (tin foil and polystyrene) in .001 and .0015 uF. Either one would do OK.

rickmcinnis wrote: ↑Mon Feb 20, 2017 7:01 pm If you come across a suitable transformer, please let me know. I have never wound one myself and figure the cost to DIY would exceed the cost of getting one already made.

These guys look to be the real deal. Nowhere near as expensive as the usual European ones, but then these aren't shielded as standard. But they do have permalloy cores with 0.16mm laminations. Thinner laminations are best as they result in lower HF losses. The primary inductance is off the scale around 230H. High inductance means a very light load at LF. But having metal cores their self-capacitance is higher than with ferrite.

https://item.taobao.com/item.htm?spm=a1 ... 5skkbbe486

abraxalito wrote: ↑Tue Feb 21, 2017 2:37 am

rickmcinnis wrote: ↑Mon Feb 20, 2017 7:01 pm If you come across a suitable transformer, please let me know. I have never wound one myself and figure the cost to DIY would exceed the cost of getting one already made.

These guys look to be the real deal. Nowhere near as expensive as the usual European ones, but then these aren't shielded as standard. But they do have permalloy cores with 0.16mm laminations. Thinner laminations are best as they result in lower HF losses. The primary inductance is off the scale around 230H. High inductance means a very light load at LF.

Does this mean not such a good LF response?

But having metal cores their self-capacitance is higher than with ferrite.

This self-capacitance relates to the LF response - does higher self-capacitance result in better LF response?

https://item.taobao.com/item.htm?spm=a1 ... 5skkbbe486

jkeny wrote: ↑Tue Feb 21, 2017 9:09 am Does this mean not such a good LF response?

Normally the opposite - higher inductance means saturation onset is moved to a lower frequency.

This self-capacitance relates to the LF response - does higher self-capacitance result in better LF response?

In my understanding the higher self-capacitance only affects the current required from the driver at higher frequencies. Nothing to do with LF.

One could disassemble them and re-arrange the laminations for lower inductance. These look like they would come apart easily.

AS recommended by Dave Slagle for his autoformers, one can tailor the inductance to suit their system.

I configured my autoformer by stacking all of the Es and Is together for a butt gap and I think the sound is better. This resulted in about ten times less inductance. The high frequencies were less confused and the bass was less floppy. My autoformers see a low load impedance.

I wish I knew lots more about transformers

Must not have looked very hard last night.

These are very good caps - replace the leads with better (copper or silver) wire.

I have used the stock silver plated steel leads and they are not awful but why use it at all?

http://www.ebay.com/itm/1200pf-5-200V-R ... 1297876915

I have bought from this fellow before. He is reliable.

Another thought - the WIMA MKP2 - 1000 and 1500 pF . Modify per Ric Schultz - use a grinder to remove the plastic on each side to remove the steel leads and solder on something you like better. Much easier than it sounds.

Off subject but related.

See this: http://6moons.com/audioreviews2/lessloss5/1.html

$90,000 dollars SD card player.

To go with the SOEKRIS DAC in the nice wooden box.

Very confusing article.

Thinking about those electrolytics on the DA-M1 - I guess the key is finding out which voltage is across them.

If it is 1.8 volts I guess they should stay (?). If 3.3 volts I think we should take them away.

I remember with the DDDAC you had to have its IIS connected or it could blow up at power on. Is this a possibility with the DA-M1? Would like to do this before actual set-up.

Hey rick, thanks for the link. The most interesting bit is in the second to last para where they indicate that the streamer only contains one clock that is used from input to output. I think that synchronicity is a step that (if DIYers can take it) will lead to significant benefits.