ISO Regen performance Improvement Cheap!

[charlesphoto]

31 minutes ago, Cornan said:

 

I am using 1mm Vanguard silver plated OCC wire. Very cheap on eBay! 

 

Regarding the LT3045. What Vf (voltage forward drop) do you use? Remember that 3 Vf is optimal for single LT3045 and 1 Vf is optimal for series LT3045. Keep any DC cable (before and after LT3045) as short as physically possible for best result. Lastly series LT3045 is significantly better than singles. Even if you combine let’s say 0.5A with 1A. Just make sure to put the highest A closest to the powered device (higher A= better)! ?

 

It’s the ebay Vanguard I’m using for my +ve - .6mm solid silver. The Kemets I ordered were the ones you recommended previously 25v/560uf. Hope that’s correct. Yes, more 3045’s but that will most likely come in the form of the MP Audio’s you are using but later this year. Of course at that point would have to redo the cable as I can’t have one of those hanging  in the air! 

 

 

[charlesphoto]

10 minutes ago, Cornan said:

All this makes me wonder if the wire gauge does’nt matter? Is it the isolation between -Ve and +Ve that matters in the end? I intend to find out! I will try my mod with a dual Supra Cat 8 wire (two of eight wires) to see if that will change anything. Hopefully tomorrow! ?

 

Yep. And it may not even matter the quality of wire so much as the wiring layout. It’s fairly early days with low voltage audio devices, so I guess everybody just assumed that better cable but in the traditional analog sense is the best way to go. Personally I’m over the moon with single strand solid wire that’s separated by air solution. Pretty sure it can be finessed even more, but for now the sound is HUGE! 

[charlesphoto]

My Kemets here next week. Ordered four of them. Should I use all four? I will solder straight to the Oyaide pins and then wire after. Or something like that depending. I think I'll look into more exotic wire for a final build once I get an MP audio power supply. An Uptone lps 1.2 probably wouldn't hurt either except for the $.

 

[charlesphoto]

4 minutes ago, BigGuy said:

Unless I am missing something, the Kemet would be in parallel regardless of where on the cap legs the wire and Oyaide tabs are attached.

Please repeat what Kemet "colors" go to v+ and v-.

 

That would be my thinking too. I just think soldering probably makes a better connection than cramming in the screw connectors. I'll try my best. 

[charlesphoto]

Just to throw another wrench into this diy dc cable build, I redid mine today using some 12awg basic romex copper hook up wire for both =ve and -ve on both ends of the lt3045 (just a stub really on the outgoing and about 4 inches on the incoming). I like it better than what I had before - more natural sounding, and better midrange. Maybe a bit less clarity/resolution. I'll give it 72 hours of burn in at least before really deciding. Ignore the wire colors on the pic! 12 gauge solid is a pain to work with - one really needs to plan out very closely the run. Would be better of course to work with some audiophile wire, but I also like free and on hand. 

 

Thinking about all of this, I've come to the conclusion that it's better to think about connection vs cable. The best of course would be short direct wire connections between the boards of the LPS, Lt3045 and mR (where the likes of the Signature Rendu probably gain an advantage). If you have to use a longer cable, then yes JSSG'ing it etc will be beneficial, but the best solution is to lose the cable altogether and just use short as possible lengths of largest gauge hook up wire. That's my theory. 

[charlesphoto]

2 hours ago, Bricki said:

Thanks @Cornan and @charlesphoto for your on going experiments. I have been trying to follow along and have had some failures with some cables sounding worse than the coax one I made last week. However, I have moved my lps1 to sit on top of my sMS-200 separated by a wooden block and isolation discs and made the shortest possible positive wire with the largest gauge cut down to fit in the screw terminals. Then the same gauge negative wire somewhat longer to keep more distance between it and the positive. I have had it burning in for 24 hours now and it makes the lps1 like a new psu. Everything is a lot cleaner but not in a washed out clinical way - cleaner in a natural and relaxed way with natural tone and texture. Annoying screach on brass instruments and some vocal pieces has been hugely reduced, restoring a more natural and far less irritating sound. There is also more depth ?. Now I'm gonna have to order some of those kemets. Where's the best place to purchase? Mouser wants to charge $24 to ship to Australia ?. 

 

‘Yep, exactly what I’.ve heard with my latest short pieces of 12awg copper. 

[charlesphoto]

Oh wow, got a listen in after 24 hours of burn in with my new cable made from 12awg romex. Sounding very very good. And that’s with basic romex wire! Just for the hell of it, I ordered some Neotech solid 12awg teflon coated upocc hook up wire for when the Kemet’s get here. Now, how I’m going to solder and fit Kemet’s and 12g wire I only thought of after the fact. I’ll make it work somehow. Otherwise I have some 14g romex in the basement!

 

I really am thinking gauge and length are big factors here. Of course one thing I could do is lose the Uptone USPCB and use a real usb cable so I could get the mR right next to the LPS-1 and do a direct connection with the 3045 inbetween. Would be way easier with solid wire esp. But then again introducing a usb cable might rise to another set of issues - and expense. 

[charlesphoto]

Kemet’s came and later today some wire etc from SonicCraft. 

 

I’m wondering why not solder the Kemet’s onto the lt3045 board - maybe the back where the screw down stand offs are attached? There’s also a slight gap I could get them into between the board and stand off, but couldn’t guarantee a good connection. It seems to me this would solve the problem of the Kemet’s coming first before the wire itself, but perhaps I’m looking at it wrong. 

[charlesphoto]

What's the best way to do that, heatsink? 

[charlesphoto]

Post doesn’t arrive until this afternoon with my final wire choice, so for the meantime shoved a couple of Kemet’s in with the wire coming in and out of the lt3045. Subtle but definitely better, particularly the top end. Seems to reduce a bit of glare. Tried a Ghent USB JSSG’d USB cable in place of the Uptone USCPB as well. Bit better high end but where’d the bass go? Back to the USPCB. 

[charlesphoto]

I'm with Cornan on this. I'm certainly thankful for any tips of course (going to use clamping tweezers between Kemet and wire when soldering for example), but I also like the freewheeling  investigative DIY nature esp. as It's just a dc cable. if the caps get screwed up and don't work, well you might hear it or might not, and it all cost pennies. Sounds worse? Snip 'em off! No need for handwringing over this. 

[charlesphoto]

So I recommend that no one tries to make this cable! It was tricky to say the least. Glad to report the Kemets never seemed to get hot, or even very warm from the soldering. Here's a couple of pics of the finished cable, and a making of. Wire is Neotech 12awg UPOCC teflon hook up. 

 

Sound is very very nice. Much more natural sounding. Will get better with a bit of burn in I'm sure. Glad I did it, but never again! (well, with solid 12awg at least). Seriously done with cable making etc for awhile. Way too much other stuff going on. Obsession officially over, but I hope the rest of you carry on! 

 

Note: there is one mistake in the first pic (wires crossed on the lt3045 input). Rectified of course. 

[charlesphoto]

10 minutes ago, sandyk said:

 

Now how about putting it all together safely in something like a small diecast aluminium box, and move the Kemet capacitors at either end to INSIDE the devices at either end ?  It may even perform a little better, too.

 

This presently looks like an accident awaiting to happen.

 

Yes, it's not much to look at, but it's actually pretty solid. I've got some heat shrink that I might clean it up a bit more with. The microRendu is in the back of the Naim DAC V1, and the LPS-1 on a shelf directly above it, right to the rear of the shelf. Unfortunately the mR juts out much further than the LPS. So it's actually all contained and nicely out of sight behind the cabinet, and the two Oyiades make a nice firm connection between the two.  

[charlesphoto]

Nice thing is I use Bridge on the mR to stream via Roon to my office set up. So by doing this one "cable" I get the better sound in both places. And it's sounding superb. 

[charlesphoto]

Oh, I did use electrical tape to cover any open joints before playing but I wanted to share a pic before that. 

[charlesphoto]

9 hours ago, sandyk said:

 

Any animals in the house ?

 It is also likely to work a little better again with the Kemets either end moved to inside the devices, especially the output end.

 

Only a fourteen year old 110 lb lab mix. He’s never shown any interest in the hifi for some reason.  

 

Keep in mind this is low voltage dc. If it was ac I would be worried about safety too. 

[charlesphoto]

Well, that’s why I’m thinking of what I’ve created as more of a hard connecter than a cable. DC Frankenconnect. And if I had to do it over again I would use 14awg instead of 12, and maybe even stranded vs solid. Haven’t had a chance to listen this morning after running in all night - too much World Cup action going on this morning. 

[charlesphoto]

Down in the office having a listen to my test album, Ahmad's Blues by Ahmad Jamal and blown away by the fruits of my creation after 24 hours of burn in. 1:30 minutes into the title track, there's a very long high note piano roll. It's always sounded like anything from a smeary trebly mess to okay. Now it sounds like an actual piano being played and one can easily distinguish each very fast key being struck. And this long time tinnitus sufferer can actually begin to tolerate the trumpet - don't get me wrong I listen mostly to jazz and think Davis, Hubbard etc are brilliant, but often it's an uncomfortable listen. Will be interesting to see a couple hundred hours from now how/if the sound changes.

 

And yeah, lets keep this fun. We're not repairing mission critical gear here. To tin or not to tin won't make that much of a difference on a dc supply I wouldn't think. 

[charlesphoto]

That’s probably a solid bet. If you planned on putting an lt3045 in-line you could probably have him make two shorter lengths with two bare ends (tinned or not). 

[charlesphoto]

2 minutes ago, fas42 said:

 

Another good example of what is gained as one evolves a system towards competency ... subjectively bad sound is a symptom of flaws in the playback chain; which can always be fixed. How one fixes it will always be a "depends" - and attenuation of problematic HF noise is usually an excellent move.

 

Well put. And it’s surprising just how far some basic mid-fi kit like the mR 1.4 and DAC V1 can be pushed once one gets the bits and pieces leading up to them right. 

[charlesphoto]

4 minutes ago, fas42 said:

 

Another good example of what is gained as one evolves a system towards competency ... subjectively bad sound is a symptom of flaws in the playback chain; which can always be fixed. How one fixes it will always be a "depends" - and attenuation of problematic HF noise is usually an excellent move.

 

Well put. And it’s surprising just how far some basic mid-fi kit like the mR 1.4 and DAC V1 can be pushed once one gets the bits and pieces leading up to them right. 

[charlesphoto]

4 minutes ago, fas42 said:

 

Another good example of what is gained as one evolves a system towards competency ... subjectively bad sound is a symptom of flaws in the playback chain; which can always be fixed. How one fixes it will always be a "depends" - and attenuation of problematic HF noise is usually an excellent move.

 

Well put. And it’s surprising just how far some basic mid-fi kit like the mR 1.4 and DAC V1 can be pushed once one gets the bits and pieces leading up to them right. 

[charlesphoto]

4 hours ago, Abtr said:

Yeah, I "wouldn't" think so too. A bad (loose) connection will introduce resistance and thus lower (DC) voltage or worse, short your DC supply. If you think "that won't make that much of a difference" then by all means go ahead and tin the wires before inserting them in screw connectors. The connections will come loose, guaranteed. One here (@bit01) complained about loose screw connections. Although this individual didn't bother to respond to my query, I'm pretty sure he/she tinned the wires. Anyway, I'm done with this. Too much trolling.  
 

 

I’m sure you are correct. My point is we don’t need to take this so seriously. Thanks for proving my point.

[charlesphoto]

Listening to my new LPs-1 to mR “cable” creation last night finally on the main system, and it just didn’t sound right. Closed down and missing high end sparkle. All the sounds were there, just not the dynamics I was used to. Lo and behold, it turns out I had upsampling in Roon ticked. Only by a power of 2, but that was enough to kill the music. I had DSP enabled as it did make a difference before this latest cable attenuating high end glare, etc. Now with the new cable it doesn’t need that. Straight through 16/44 sounds best. Similar to what I’ve experienced comparing 16/44 with MQA. 

[charlesphoto]

Very nice Paul. In the end I wish I had just soldered the caps to the board like you did. The halfway up the leg thing was really difficult, and I'm just not sure about those tiny splines in the screw downs vs the whole wire. I guess the proof is in the listening. Did you add the heatsinks so that you could heat shrink it?