The true experimental tweak thread

[lmitche]

24 minutes ago, Cornan said:

 

Nice work! Great to hear that it improved things! ? I think I will try it on my Louren Fisher hifi stand as well. It doesn't have special grounding screws, but just grounding the metal tubes should work as well to reduce EMI/RFI.

I don't know anyone who is as well grounded as you!

[lmitche]

4 hours ago, BigGuy said:

 

Thanks as always, Cornan.  That was quick and looks like a nice replacement for what I have.

 

       Question from the rookie...  My existing regulator board  <https://www.ebay.com/itm/AC-Input-TPS7A4700-Ultra-Low-Noise-Adjustable-1-4V-20-5V-1A-Power-Supply-/391736075888?hash=item5b354a2670>

has a Noise spec of 4uM vs the <40uV of the LT1963 based board.

       This is the existing xformer used with the "4700" board...

< http://www.newark.com/multicomp/mcta050-12/toroidal-transformer/dp/38K4881?ost=mcta050%2F12&iscrfnonsku=false&ddkey=http%3Aen-US%2FElement14_US%2Fsearch>

       In the greater scheme of things, what is generally considered "ultra low noise" and how much of a difference does a 10x factor really make?

Hey BigGuy, what happened to the tps7a4700 board? Sounds like a great board.  Why wouldn't you want another?

 

I have a spare 1963 board in a metal box. We should talk.

 

Larry

[lmitche]

11 hours ago, sandyk said:

 

 The LT3045 1A PCB would be dropping too much voltage (7v at 1A = 7W of heat to dissipate) when used on it's own to provide +5V from an internal +12V supply rail, so would not be suitable for use internally.

Even the more typical current draw of around 500mA from an SSD would see it dissipating 3.5W of heat.

The ebay 40uV regulator has a decent sized heatsink,  so it is suitable for use from a PC's internal +12V rail.

Using an external PSU is not always a good idea, as it gives the possibility of an earth loop being created, whereas using the ebay 40uV regulator sees it using the same 0 V ("earth") reference point in the PC itself.

Incidentally, several C.A. members have used the 40uV ebay PSU that I previously told them about, and reported back favourably.

You can also use the rectifier diodes in this PCB to drop a little more voltage, and thus run a little cooler, as well as gain a small amount of additional filtering. When used this way, it may also be a further advantage to connect something like a 100uF electrolytic capacitor across the A.C. input terminals. Make sure you get the polarity of the capacitor correct though.

I will see if I can dig up the 40uV's  schematic to illustrate this point.

Regards

Alex

I have one of the eBay 1963 regulators with heatsink dropping 19 volts to 13 followed by a 1 amp 12 volt lt3045 into a wifi adapter and Netgear switch. Works great!

[lmitche]

33 minutes ago, BigGuy said:

Appreciate the evaluation by someone obviously knowing his way around power supplies and regulators.

 

I may opt for the S12 despite its being more expensive.

 

Thanks!

I have three Sigma 11 based LPSUs here one at 12 volts and two at 19 volts.  They are just fantastic.  You can buy premade sigma 11 regulators at ymb.com. Integration with a case, toroid and iec and dc connectors isn't too bad if you are handy with drilling metal.

[lmitche]

5 hours ago, TubeMan said:

https://www.ebay.com/itm/100VA-Ultra-low-Noise-100W-LPS-R-core-Linear-power-supply-DC-5V-24V-With-display/131867485038?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2055119.m1438.l2649

 

 

Yes, this may be good, but it's not using the original sigma 11 board design shown here: https://www.amb.org/audio/sigma11/

[lmitche]

6 minutes ago, BigGuy said:

Good to hear about the Sigma 11 LPS.  BTW, site is <ymbaudio.com>  Would be nice if the S12 was as good. 

 

May I ask what approx did you pay for the Sigmas?

It wasn't a commercial deal, just a deal between friends, sorry.

 

Yes, my mistake, ymbaudio.com is right.  Amb.org is the diy site for the sigma 11.

 

What's a s12?

[lmitche]

2 hours ago, jean-michel6 said:

Hi All ,

 

Does anybody has tried the LT 3045 regulator board with either an SPS-500 or a LPS-1 ?

 

To power what and What was the results ?

 

Thank's

Here two lps-1s 7 volt outputs power three ldovr.com lt3045 6 to 5 volt chains.  One is used to energize a USB 3 pcie card. The second y-cable chain energizes an ISO  Regen and injects power into my DAC.  SQ is truly fantastic!

[lmitche]

1 minute ago, lmitche said:

Deleted

[lmitche]

3 hours ago, Balázs said:

Okay let's summarize it:

 

11 VAC --> ^[LT1963] --> *7 VDC --> ^^[LT3045] --> **6.3 VDC

 

^   the LT1963 has got a bridge rectifier made of HER303 diodes as shown on the picture you included

* <40uV low noise DC stage one, LT1963 PCB set to 7Vdc 

^^ Schottky bridge rectifier dropping 0.4 or 0.8V 

** <0.8uV ultra-low noise DC stage two, LT3045 PCB set to 6.3Vdc

 

voila - we've got it

 

 

I have several of these LT1963/LT3045 chains here and they run cool and sound great.

[lmitche]

4 hours ago, Balázs said:

Thanks @lmitche good to know I'm walking the right way would you mind sharing some more details like

 

-what are your input/output voltages

-what are your temperatures

-are you using more of these boards in series (cascaded) 

 

Cheers

AC(120v) >Sigma 11(19v) > lt1963(13v) >lt3045(12v) >ycable(leg 1) >Netgear fs108 v3

                                                                            >ycable(leg 2) >lt1963(7v) >lt3045(6v) >lt3045(5v) >TRENDnet 4-Port Broadband Router, 4 x 10/100 ports

 

I don't have any temps as everything is cool to the touch.

 

 

[lmitche]

4 hours ago, Cornan said:

 

Yes, that is really a firm conclusion from my part as well. Also a little question mark for the use of a ground shunted SMPS from my side. IME JSGT is only beneficial for ethernet devices usch as network switches and routers. Everywhere else it actually descrease SQ in my setup. In my setup the SQwith a different feeder supply was’nt  minor either. I cannot wait to see how the HDPULN LT3045/3A single ps) between Gophert and LPS-1.2 turns out and ofcourse how it turns out with one more post LPS-1.2. More about this later though! ?

Agreed, except that the JSGT on the 12 volt input of my USB 3.1 HDD enclosure benefits SQ as well.

[lmitche]

19 minutes ago, Cornan said:

 

Interesting! Makes me wonder if a HDD have high impedance leakage  just as a network switch? That would explain a lot of the sensitive nature of HDDs. I wonder how a floating power supply turns out powering a HDD? ?

Perhaps it's something to do with the electric motor or actuator.

[lmitche]

7 minutes ago, Cornan said:

 

Yeah, and the back is as gorgious as the front! ?

I immediately change my order from one to two single ps. I still dont know how much they are...but I dont care! ?

 

 

 

 

Michael does a great job of finding heatsinks that look like sculpture. The bluish one on my 3 amp model made it impossible to hide in a box.

 

I wonder where he finds them.

[lmitche]

2 minutes ago, Cornan said:

 

It is Stammheims brand new 5A LT3045 single ps. I have ordered two to replace my 3A LT3045 pre my Brooklyn DAC. I will use the 3A single ps boards for my planned LPS-1.2 project explained earlier on this thread.

It's an AC to DC regulator as shown, but can be ordered DC to DC only.

[lmitche]

4 hours ago, jaaptina said:

LPS-1 is 7,5V so I chose 7, 1V and 6,0V. Could be the first step is to small. But the problem occurs since adding the 6,0V LT3045. 

 

Thanks for the suggestions. I'll give it a try. 

Lps-1 output is 7 not 7.5 volts. A vdrop of zero is too small. Try removing the 7 volt lt3045 board.

[lmitche]

8 hours ago, Cornan said:

Three georgious looking 5A LT3045 boards just passed my door step! ?

 

 

A pity I have’nt got time to start with them until the weekend. Love at first sight though! ?

Wow, Cornan are you a professional photographer? How about a picture of the bottom?

[lmitche]

45 minutes ago, Cornan said:

 

Great! ? Make sure to use your best power supply for the BluWave Spdif output. Gophert>HPULNs>LPS-1.2 (5v) really makes BluWave shine at that spot. I might even try my last HC-HPULN there instead of using it to power my Aries Mini. ?

Cornan, You may want to try the 3 amp lt3045s after the lps-1.2, not before. I'd add one, listen , and then add another. Larry

[lmitche]

5 minutes ago, Cornan said:

There seem to be a lot to gain from a good feeder supply for LPS-1.2.

 

Those are very good words, reflect long and hard about them my friend!

[lmitche]

5 hours ago, [email protected] said:

Hi Cornan

1st 

with 15vac you will get 20+VDC after rectifier!!! And if you want to have 12VDC out  @3A - need to dissipate 24W!!!

2nd

i am not agree with PSRR calculation for the 2 reg in series

Cornan, Alexey is right to point out you may smoke your boards with a 15 volt ac transformer. The voltage of the rectifiers are likely to be over the maximum input thresholds of the lt3045s.

[lmitche]

http://www.bristolwatch.com/ele/basic_ac_rectification.htm

[lmitche]

5 hours ago, Cornan said:

 

I would actually be extremely surprised if 15VAC would smoke the HC-HPULNS since they are still suitable for 20V plus capable to endure a higher voltage drop and current than other LT3045s. However, just in case I have sent Michael a e-mail to make 100% sure that I will be fine. I will let you know his answer.

My intension will be like this: 15VAC - 20VDC>15VDC - 15DVC>12VDC - Brooklyn DAC (plus possibly additional aluminium polymer cap inbetween the HC-HPULN)

 

Happy to hear you are reaching out to Michael. He will steer you the right way.  He did the same for me.

[lmitche]

4 minutes ago, Cornan said:

 

I don´t actually know until I have tried. There are several advantages with the AC-AC PS as a feeder supply compared to the Gophert. Things like balanced transformers, EMI input filter, torodial transformer, bigger transformer, no pre-regulation (=possibly less noise) etc are things that I consider an advantage to the Gophert. Both are floating. Gophert have an advantage of optimal pre-regulation (= possibly less heat). Other have said (over and over again) that the SMPS design is Gophert´s culpit...but I do not agree that it is true in my perticular setup!  

Anyway, the reason I bought the AC-AC PS is to find out if it could concur my Gophert? I really do think it could, but mainly due to the things I mensioned above and not due to the actual SMPS design. The balanced & floating design is the most important apect IMO.

You could add a $35 12 volt Sigma 11 discrete regulator board between the AC transformer box and the lt3045s and have a kick ass 12 volt solution. That's what I did here with Michaels 3 amp lt3045 board.

[lmitche]

4 minutes ago, Cornan said:

 

Interesting! Do you have a link to that Sigma 11 discrete regulator board?

Here you go!

 

https://www.ebay.com/itm/S12-Super-regulated-linear-power-supply-board-5-30V-LPS-10000uF-50V-Versio-PSU/132048787951?hash=item1ebeb899ef:g:odUAAOSwbopZPMnE

 

But on second thought, a 15 volt transformer may not be enough. You may need 18 volts AC into a chinese Sigma 11 to get 12 volts out.

 

Another approach is to find an lt1083/1084 board with enough current for your application. Here is one:

 

https://www.ebay.com/itm/LT1083-7A-Adjustable-Linear-Voltage-Regulated-Power-Supply-Board-Self-Recovery/282892603399?epid=2130705681&hash=item41ddb69007:g:3ucAAOSwutNasRD0

[lmitche]

I did some sleuthing and the Sigma 11 should work with your 15 volt ac transformer Just get one set for 11 or 12 volts dc output and run into Michaels boards.

[lmitche]

35 minutes ago, Cornan said:

 

No links available since it is still not an official release yet. It is the same one as Stammheims 3A LT3045 boards, but 5A and with Kemet A750 output caps. It is a AC-DC PS board that I use in series.

Yes, but in this case the sigma 11 will do the AC to DC conversion and the two hc-hpuln boards will be using DC on their inputs.