• These commercial threads are for private transactions. diyAudio.com provides these forums for the convenience of our members, but makes no warranty nor assumes any responsibility. We do not vet any members, use of this facility is at your own risk. Customers can post any issues in those threads as long as it is done in a civil manner. All diyAudio rules about conduct apply and will be enforced.

FS: „Ultra Low Noise“ Power Supply, LT3045 based PCB’s

Jump to Latest
S

Stammheim

Member
Joined 2012
mpaudio.net
Hi guys,

sorry for late response, I visited the "Highend" in Munich yesterday and came back now.
Please notice, that the LT3045's maximum input votage is 20VDC, while the maximum output voltage is 15VDC. Don't use voltages above these figures.

Use following formula to get DC after rectifying:

trafo AC x 1,414 - Vf (diode's voltage drop of 0,4V in this case)

You will see that your trafo's AC voltage is 14VAC max, matching the LT3045's maximum input voltage.
More important is get the right voltage matching your needs (with formula above).

get-lit said:
I've been reading up on transformer EMI, related to DACs and power supplies. It seems there's general consensus Toroidal transformers have low enough EMI to not be of significance even in the same enclosure as the other PS components, and any advantage in remote placement is more than offset by disadvantages of loop area and impedance, wherein it's generally best to keep everything wired closely. A shielded transformer seems to be more than adequate. Of course this is only a line of thought, but it's widely used, so I'm leaning toward a single self-contained rack unit with shielded transformers. DAC transformers also tend to be low VA rated anyhow. To be sure, I may as well add a second shield between the transformer enclosures and other PS components. I'll include EMI absorption on one plane of the transformer enclosures and on the reflecting planes of the air vent ports. DC fans still powered separately.
Click to expand...

Thats it!!!
Keep wires short, especially the transformer-AC-wires. I recommend a shielded trafo, if you feel better use a EMI / copper shield between trafo and your devices.

Best regards

Stammheim
 
G

get-lit

Member
Joined 2009
Thanks Stammheim, I keep forgetting about the voltage range.

Cornan, I'm really interested in hearing how your series HC-HPULN turns out without the Sigma 11 based PS. I would like to be able to eliminate it if it's not much different without it. It's more energy consuming, more heat, and I "presume" less efficient at filtering than LT3045 based PS. I'd like to compare by specification but unfortunately I'm unable to find ripple rejection for the Sigma 11. Also the HC-HPULN having 6x parallel LT3045 with caps will have measurably greater ripple rejection than the single LT3045.

Also you could get the 1/2" 6061 for $20 more, although it probably wouldn't make any difference. Also, most these metals only have <9% difference EMI blocking effectiveness among them, so which you choose is not very critical.
 
D

Daudio1

Member
Joined 2012
get-lit said:
interested in hearing how your series HC-HPULN turns out without the Sigma 11 based PS. I would like to be able to eliminate it if it's not much different without it. It's more energy consuming, more heat, and I "presume" less efficient at filtering than LT3045 based PS. I'd like to compare by specification but unfortunately I'm unable to find ripple rejection for the Sigma 11.
Click to expand...
Sigma 11 is very good regulator, but don't know specs. Think single digit mv noise. Probably online at AMB site, but navigation there takes some effort.

I like the idea of LT3045 modules as post-regulators, where the 1st regulator does the heavy lifting, passing a fairly clean DC on to the post-reg, where it can remove the small remaining noise more easily. Can't prove that, but how it lays out in my head. Don't think the pre-reg needs to be very low noise, but wouldn't use LM317
icon12.gif
In fact, working on LPSU with cheap asian LT1963 module as pre-reg and LT3045 module as post-reg (maybe 2 ?).
Might be interesting for you to test various combos ?

re; shielding - for magnetic field not EMI. Don't forget better shields in air(distance) and mild steel.
 
Cornan

Cornan

Member
Joined 2017
get-lit said:
Thanks Stammheim, I keep forgetting about the voltage range.

Cornan, I'm really interested in hearing how your series HC-HPULN turns out without the Sigma 11 based PS. I would like to be able to eliminate it if it's not much different without it. It's more energy consuming, more heat, and I "presume" less efficient at filtering than LT3045 based PS. I'd like to compare by specification but unfortunately I'm unable to find ripple rejection for the Sigma 11. Also the HC-HPULN having 6x parallel LT3045 with caps will have measurably greater ripple rejection than the single LT3045.

Also you could get the 1/2" 6061 for $20 more, although it probably wouldn't make any difference. Also, most these metals only have <9% difference EMI blocking effectiveness among them, so which you choose is not very critical.
Click to expand...

I will surely report back here! :) A couple of notes though. I will be using the HC-HPULNs which is 5A out and 10x parallel LT3045 without Sigma 11 and the HPULN which is 3A out and 6x parallel LT3045 with Sigma 11 (with a 1A LT3045 board in series).

Heat is my biggest concern, but my Brooklyn DAC does’nt really pull more than 1.5A during start-up and below 1A during music listening. The Sigma 11 will be powering low current devices (between 0.2-0.8A) and since I is the voltage reg that will get the largest voltage drop I will probably mount it inside the balanced isolated ps onto the heatsinks on the side wall (hopefully it will fit). The LT3045s will be outside without enclosures as I want them to be.

Since I use a Gophert csp-3205II to power the HC-HPULNs at the moment I have actually three voltage regs in series right now (Gophert>HC-HPULN>HC-HPULN) even if the first voltage reg is certainly not that great. The balanced ps have many other advantages to the Gophert that I do think will overcome that advantage, but it is good to keep that in mind when I start comparing the two.

You are probably right that the different aluminum plates does’nt make a huge difference if you are not aiming for the best. However, the 7075 should be avoided according to J.S.
 
G

get-lit

Member
Joined 2009
I'm trying to help LittleScarabee determine a suitable transformer for his DAC requiring 12VDC input 0.5A.

Assuming he's using single L-HPULN or HC-HPULN without any other first stage PS, he may as well go with 3V voltage drop because he's only at 0.5A.

From what I understand...
TransformerVoltagePeak = PowerSupplyVoltageOut + PowerSupplyVoltageDrop + RectifierVoltageDrop

RectifierVoltageDrop = 1.5V (typical)


Substitute...
TransformerVoltagePeak = 12V + 3V + 1.5V = 16.5V

Determine TransformerVoltageRMS...
TransformerVoltagePeak = TransformerVoltageRMS x 1.41

Substitute for TransformerVoltageRMS...
TransformerVoltageRMS x 1.41 = 16.5V
TransformerVoltageRMS = 16.5V / 1.41 = 11.7V

Therefore, he'd need a 12V transformer.

Then VA rating would be...
TransformerVoltageRMS x Current = 12V x 0.5A = 6VA
I'm not sure but I think you're supposed to add a percentage for headroom, so maybe go with 10VA?

Consumed energy or heat at the L-HPULN / HC-HPULN would be...
PowerSupplyVoltageDrop x Current = 3V x 0.5A = 1.5W

Seem right?
 
Cornan

Cornan

Member
Joined 2017
I'm not sure but I think you're supposed to add a percentage for headroom, so maybe go with 10VA?
Click to expand...

This is the rule of thumbs to calculate VA of transformer:
Loads watts/0,6= VA of the transformer
12v*0,5A=6/0.6= 10 so IMO 10VA is sufficient as long as the peak A is not more than 0,5A.
Everything else looks good. Personally I will be using a 12VAC 50VA transformer for my Brooklyn DAC that have a peak A of 1,5A so I am playing very safe (only need 30VA)! ;)
 
Last edited:
Cornan

Cornan

Member
Joined 2017
Stammheim said:
Hi guys,

regarding the diode's Vf (voltage drop): Take notice, that the used schottky has a drop of about 0,4V only.
If place is not the topic, a trafo with more VA then necessary is always to be preffered. Especially, because projects change and the need of a "bigger" VA trafo is always possible.

Best regards

Stammheim
Click to expand...

Is there something like rectifier diods that will drop ~1,5A? If there is I am interested in a HC-HPULN (5A) with rectifiers like this. It will ensure that my other HC-HPULN in-line will perform optimal in my coming balanced isolated LPSU.
 
D

Daudio1

Member
Joined 2012
Stammheim said:
You can use other diodes, like the HFA's (~1,4V Vf depending on If). Great diodes btw. The problem with higher drop is more heat. I measured about 95C° and more with these diodes, so I would recommend them with a less current need.[/QUOT]

Or with heatsinks. He could also use 3 or 4 regular diodes in series to get a higher total Vf to tame his voltage problem.
Click to expand...
 
Cornan

Cornan

Member
Joined 2017
Stammheim said:
Cornan, I guess you mean ~1,5V drop.
You can use other diodes, like the HFA's (~1,4V Vf depending on If). Great diodes btw. The problem with higher drop is more heat. I measured about 95C° and more with these diodes, so I would recommend them with a less current need.

Best regards

Stammheim
Click to expand...

Thanks a lot, I will keep the HFA´s in mind if needed! :)
I will only have a peak A of 1.5A and continuesly around 0,8-1A with 12VDC so I really think it will be alright. Right now I will need to have a 2,3V drop-down between the HC-HPULNs, but I would prefer to keep it a bit lower. With the HFA´s or perpahps this rectifier bridge (asking the eBay seller right now about the voltage drop-down) HPOO Single power supply rectifier filtering finished board (NOVER Version) | eBay
I will end up with 12VAC*1,414(-0,4)-1,4=15,2-15,6VDC and roughly 1,5-1.7V drop-down which is pretty much where I want to end up.
 
Last edited:
Cornan

Cornan

Member
Joined 2017
Daudio1 said:
Stammheim said:
You can use other diodes, like the HFA's (~1,4V Vf depending on If). Great diodes btw. The problem with higher drop is more heat. I measured about 95C° and more with these diodes, so I would recommend them with a less current need.[/QUOT]

Or with heatsinks. He could also use 3 or 4 regular diodes in series to get a higher total Vf to tame his voltage problem.
Click to expand...

As I recently posted to Stammheim I am looking into adding this pre the first HC-HPULN HPOO Single power supply rectifier filtering finished board (NOVER Version) | eBay

s-l1600.jpg

s-l1600.jpg


I have send a pm to the seller for info regarding voltage drop down of this perticular rectifier bridge.
Click to expand...
 
Last edited:
Cornan

Cornan

Member
Joined 2017
Daudio1 said:
Cornan said:
Eeeeeuu, ugly high noise bridge rectifer :( Don't you want quiet Schottkys or Ultra-fast-recovery diodes for your audiophile LPSU ?
Click to expand...
*LOL* yes infact I do! :D Rectifier diods is out of my comfort zone, so I could surely use some help picking out the ultimate one. Do you know of any rectifier bridges that you can recommend that drop 1,4-1,5V additional to the one included in HC-HPULN (-0.4v)?
Click to expand...
 
jean-paul

jean-paul

Disabled Account
Joined 2002
Daudio1 said:
Or with heatsinks. He could also use 3 or 4 regular diodes in series to get a higher total Vf to tame his voltage problem.
Click to expand...

When you build an ULDO with rectifier diodes in series or rectifier diodes with higher Vf for higher voltage drop then things do not correlate :)

The "voltage problem" is self inflicted and can be solved an easier way. It would be nice to see the measured difference between a single PSU with correctly chosen transformer and the various regs in series + higher voltage drop diodes situation. In many cases simpler is better but this is, of course, a general observation.
 
Last edited:
D

Daudio1

Member
Joined 2012
Cornan said:
Daudio1 said:
*LOL* yes in fact I do! :D Rectifier diodes is out of my comfort zone, so I could surely use some help picking out the ultimate one. Do you know of any rectifier bridges that you can recommend that drop 1,4-1,5V additional to the one included in HC-HPULN (-0.4v)?
Click to expand...

Stammheim already recommend some previous in thread, and uses good ones in products!
I searched Mouser for 25V 3A Schottky diodes to replace more ordinary types in a LT1693 module to upgrade with R-core, JS snubber, and LT3045 post- reg. So not hard to find, and cheap. Will find part number if asked, but don't know Vf, easy to find in part specs pdf (link on mouser item page)

jean-paul said:
... then things do not correlate
... In many cases simpler is better ...
Click to expand...
What mean 'not correlate' ?
I only suggest as workaround to situation, and agree about simplicity.
Cornan has great drive, energy and unique style in audio experimentation. Not want to suppress that. Just maybe a little nudge here and there to point to alternate and/or beneficial paths
Click to expand...
 
Last edited:
jean-paul

jean-paul

Disabled Account
Joined 2002
Sorry, 25V rated Schottky diodes are NOT the right diodes for a 12V transformer (if we are still dealing with a 12V transformer). Please do the math and use peak-peak values for voltage. I would choose 40V or even higher rated Schottky diodes for a 12V transformer...

You are right about enthusiam and empirical findings being OK but the KISS principle and oldfashioned calculations are still valid :)
 
Last edited:
You must log in or register to reply here.
Top Bottom