Go Back   Home > Forums > Amplifiers > Power Supplies

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 7th November 2013, 04:47 PM   #1
SGK is offline SGK
diyAudio Member
 
Join Date: Dec 2012
Location: London
Default Help a novice steadily build a linear 12V power supply?

Hi

Hopefully this little project by a novice is worthy of your attention and assistance!

I have assembled an audio server made up with an ASRock Z87E-ITX mobo, Haswell 4570T processor, ESI Juli@ XTe sound card with clocks upgraded with a Fidelity Audio Micro Clock 2 and PicoPSU power supply with 12V 'brick'.

In the interest of achieving a better power supply and learning more about electronics I have been upgrading the power supply. As a first step, and with a lot of kind assistance from Brent of Fidelity Audio here in the UK, I assembled with point-to-point wiring a 12V 'rail' comprising two 22,000uF Fidelity Audio 'SI' caps, an SPower HC 12v regulator and another 22,000uF 'SI' capacitor, fed 18V from another 'brick'. The caps and regulator may well be overkill for this application but it seems to work very well. This 12V powers the PicoPSU and the Micro Clock 2.

I'm interested in taking this a step or two further. One direction, although it may not be the first one that I implement, is to replace the brick with a 160VA toroidal transformer with 15V secondaries and a full-wave bridge rectifier made from four Cree C3D08060A Silicon Carbide Schottky diodes. 160VA may well be more than enough but it fits.

The second direction involves what I believe is a CLC filter. Specifically, it has been suggested to me that my existing setup could be improved further by adding a series of film capacitors and a choke ahead of the 'SI' caps:

Input
Wima MKP10 0.47uF 1600V
Wima MKP10 0.047uF 1000V
Wima MKP10 4700pF 1000V
Coilcraft CMT4 26mH 6A 0.115Ohm
Existing caps...etc

I picked up these parts (I got the choke cheekily as a request for sample and the Wima caps were essentially free as they tipped my Mouser order into free delivery) but as I read a bit more about CLC filters I'm unsure that these cap values are appropriate for this application or indeed whether the filter adds anything beyond the existing cap/regulator setup. (The choke is, ahem, huge but can be made to fit.)

I'd therefore like to start by asking for thoughts on the filter, both with respect to adding it into my existing 'brick' setup and with respect to the transformer/rectifier setup.

I'm sure I will have questions on the transformer/rectifier as well at some point.

Thanks in advance

Steve

Last edited by SGK; 8th November 2013 at 02:59 PM.
  Reply With Quote
Old 8th November 2013, 02:43 PM   #2
diyAudio Member
 
Join Date: May 2011
Location: Silicon Valley
I suggest that an excellent application of the next ~ 239 of your DIY budget, would be to purchase an oscilloscope with probes. Then you can view the voltage waveforms at the various nodes in your existing power supply, to see how well or poorly it is performing. You can also discover anomalies due to poor ground layout and/or inefficient go-and-return loop wiring. You can study the amount of ripple, and watch as it decreases (or fails to decrease!) as you march your probe from the transformer to the output. You can study noise and compare against theoretical predictions from ideal lowpass filter equations.

Finally, as { diyAudio member zigzagflux } points out, when you have an oscilloscope, you can use it to dial in an optimum snubber, which completely damps any possibility of transformer secondary ringing.
  Reply With Quote
Old 8th November 2013, 03:10 PM   #3
SGK is offline SGK
diyAudio Member
 
Join Date: Dec 2012
Location: London
Thanks for the response! I guess I might have seen that one coming. :-) In the immediate short term, I can likely borrow one from a good friend and learn a little about how to use it - assuming I can get him interested in my project (he is relatively new to electronics but has made the investment you recommend).

As I read and tried to absorb the discussion in the snubber thread and your Quasimodo thread (and the material you linked to) I was thinking the issue of transformer secondary ringing and tackling that with an optimum snubber was not necessarily the same thing as what the CLC filter which was proposed to me (as described above) or rather that these were separate issues (as separate as things can be in electronics). Is wrong?

I suspect you would tell me to get the toroid and rectifier up and running and then work on optimising that setup as I learn more rather than trying to improve on a brick DC input which will ultimately be replaced. To that end, can I check that I have properly interpreted the pin configuration of the Cree diodes into an actual wiring diagram for the rectifier? Apologies for the corny graphic!

Click the image to open in full size.

Thanks again

Steve

Last edited by SGK; 8th November 2013 at 03:16 PM.
  Reply With Quote
Old 8th November 2013, 03:50 PM   #4
diyAudio Member
 
Join Date: May 2011
Location: Silicon Valley
Your drawing looks correct except the ~L wire is labeled "A" at lower right. Also notice that the case of each diode is electrically connected to pin 1, so if you mount the four diodes on a common heatsink, you'll need electrically insulating (but thermally conductive!) mica tabs + thermal grease for assembly.

You can download a copy of the excellent LTSPICE circuit simulator; it's free of charge and available at Linear Technology's website. If and when you put your circuit design into LTSPICE and simulate it, I predict you will discover that Brent_of_Fidelity_Audio's CLC filter does not snub ~ 0.3 Megahertz transformer ringing. Instead, I predict you will discover that the CLC filter reduces 120 Hertz ripple that occurs when the diodes recharge the filter capacitors, and the (output) load current discharges them, on each half-cycle of the 60Hz mains frequency. But why take my word for it? Simulate it yourself and make up your own mind. Remember, LTSPICE is free.
  Reply With Quote
Old 8th November 2013, 04:31 PM   #5
SGK is offline SGK
diyAudio Member
 
Join Date: Dec 2012
Location: London
Thanks. Noted re A vs L. On the mounting, I have TO-220 mounting kits with the isolation grommets and also have Sil-Pad 2000 pads (rather than use the mica pads and thermal grease). As regards heat sink, I was hoping to merely use the bottom of my enclosure (2.5/3mm aluminium) but I am unsure of what heat I can expect/need to cater for.

Re LTSPICE, I will check it out. Hopefully it runs on a Mac. (I now see that it does.)

I should be clear that the Wima film caps and choke were not recommended by Brent but rather another newbie but more experienced than me DIYer. Probing the thinking around caps values and effectiveness is why I eventually took my questions here. Brent guided me solely in relation to adding, on my inquiry, the SI caps and regulator and higher voltage brick versus the brick 12V input that comes with the PicoPSU. I'm trying to take that to the next stage.

BTW what should I look out for in an oscillator and do you have any brand and model to recommend? My friend has a rather more expensive Bitscope.

As regards using it to check ripple, I will browse the web over the weekend but, in short, is it a case of examining Fast Fourier Transforms and identifying spikes?

(I should add that I am in the UK)

Last edited by SGK; 8th November 2013 at 04:41 PM.
  Reply With Quote
Old 8th November 2013, 05:00 PM   #6
Mooly is offline Mooly  United Kingdom
diyAudio Moderator
 
Mooly's Avatar
 
Join Date: Sep 2007
If you have any doubts on connections etc (and even if you don't) I would recommend using a bulb tester (search the forums ) in the primary. It will save you from any expensive mistakes.
__________________
-------------------------------------------------------
A simulation free zone. Design it, build it, test it.
  Reply With Quote
Old 8th November 2013, 05:25 PM   #7
SGK is offline SGK
diyAudio Member
 
Join Date: Dec 2012
Location: London
Wise advice - thanks.

Yet another question. I was thinking of using a magnetic hydraulic circuit breaker rather than a fuse. The question is what rating. I read the material on Rod Elliot's site. There he outlines a process for determining an appropriate fuse level. If I have read the material correctly, it suggests a fuse of no more than half (safety factor of 2) the primary voltage divided by the short circuit impedance of the transformer (Rsc). In turn, Rsc is given by the short circuit voltage (Vsc) of the transformer divided by the full load current (IFL).

For my 160VA transformer with a primary voltage of 230V, IFL = 0.7A.

The challenge is figuring out Vsc. The higher Vsc the lower the fuse rating required to protect the transformer. I think I am being conservative assuming a Vsc equal to the secondary voltage in which case Rsc = 15/0.7 = 21.4 and 230/21.4 = 10.7 which in turn implies a 5A fuse/circuit should be enough to protect the transformer. 5A should also be more than enough for the power supply. I could probably go lower but I am not sure it is necessary.

Am I thinking along the right lines?
  Reply With Quote
Old 8th November 2013, 05:42 PM   #8
Mooly is offline Mooly  United Kingdom
diyAudio Moderator
 
Mooly's Avatar
 
Join Date: Sep 2007
I'll be honest and say I've no experience of using anything other than conventional protection (fuses etc) for mains transformers. One factor to remember is that a toroidal can draw a large switch on surge seemingly out of all proportion to its steady state idling current.

Empirically... I would have suggested a slow blow fuse (T time delay or AS anti surge rated) of around 2 or 2.5 A for a 160 va toroid.

For full protection the secondaries should be fused too such that if a fault such as a shorted load or rectifier etc occurred then the secondary fuse would/should probably blow before the primary.

A lot depends on the characteristics of the transformer. There's no real cast in stone right and wrong answer to this. I once had a Pioneer 150w RMS per channel amp and that had only a primary fuse. There was no fusing at all on the secondaries but it was a conventional E I transformer rather than a toroid with its huge inrush currents. So I guess in that case the primary fuse could be rated accordingly. A diy amp I built using a 500va toroid needed a T6.3 amp primary fuse to prevent nuisance blowing on switch on.
__________________
-------------------------------------------------------
A simulation free zone. Design it, build it, test it.
  Reply With Quote
Old 8th November 2013, 07:47 PM   #9
SGK is offline SGK
diyAudio Member
 
Join Date: Dec 2012
Location: London
Determining the appropriate amps rating for each should be the same, no?

I guess one thing I'm unsure of is relating amps on the primary to amps on the secondary. I'd like to have plenty of headroom in my 12V supply - say 5A. I figured that I needed 15V secondary on the transformer on the basis of the rough calc 15 x 1.4 -2 = 19V and I have a 5V dropout with the SPower regulator.

Is it appropriate to relate wattage on the primary side to wattage on the secondary? That is, 2A x 230V = 460W which is plenty of headroom for 5A x 15V = 75W for the rail? I didn't think this was the case until I learnt that the fuse in my large toroidal transformer based linear power supply in my Oppo 103 mod (Dr Lee) could be replaced with a 2A magnetic hydraulic circuit breaker.

EDIT: a little more reading and I think I can answer my own question. The power delivered to the secondary is equal to the power delivered to the primary less any loss due to inefficiency (e.g.heat) in the transformer itself. So a 2A fuse should be more than enough.

Last edited by SGK; 8th November 2013 at 08:00 PM.
  Reply With Quote
Old 8th November 2013, 09:52 PM   #10
SGK is offline SGK
diyAudio Member
 
Join Date: Dec 2012
Location: London
Well I spent a couple of hours poking around the Mac version of LTSPICE and I must say I am hopelessly lost. The introduction guide seems to be written for the Windows version as I don't even see the same menus/toolbars. I managed to draw a basic circuit with a few wires and components but that's about it. Clearly I have a lot to learn before I am in a position to run any simulations.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
WTB F5 power 12V input smps power supply JZatopa Swap Meet 2 19th June 2011 01:41 AM
How do i build a 12v power supply ? onis_uk Power Supplies 15 3rd October 2010 11:52 AM
Power One 12V 10A linear power supply boywonder Swap Meet 0 11th September 2010 03:38 PM
12V Power Supply ? samsagaz Power Supplies 10 4th June 2005 05:38 PM
Looking for transformers to build +5V, +12V power supply for CD-ROM drive Saurav Parts 4 19th September 2003 02:21 PM


New To Site? Need Help?

All times are GMT. The time now is 04:58 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2