| ezkcdude |
| As the topic says, I'm going to use a 16VAC wall wart to power a DAC. The DAC will have onboard regulation using LM317/337. Basically, I'm doing this to avoid dealing directly with transformers and mains power, as a safety issue. After the rectifier, is it enough to use one large electrolytic (4700 uF) to smooth the ripple before the regulators? What would you guys do in my situation? |
|
|
| cliff |
How do you plan to get +/- supplies out of a single AC winding?
What voltages do you want to end up with?
At what currents?
Are they balanced? |
|
|
| ezkcdude |
I thought the +/- is supplied directly from the wall wart coaxial plug. It came with a schematic of the transformer, but I don't have it on me right now. I'll check tonight.
As for the voltages, I want to end up with +/- 12, +5, +3.3 for the various sections. The current draw should be well under 1 A, but I can't say for sure right now. |
|
|
| AndrewT |
Hi,
using about 2mF per A is usually enough for accepatable ripple before the reg.
Your 4m7F should easily cope.
There was a thread recently where a single AC winding was purportedly giving +-DC output. I am not sure the poster knew what he was proposing, worth searching for and try to fathom his schematic. |
|
|
| ezkcdude |
| Now I understand what I proposed doesn't make sense. I just don't want to mess with transformers. Well, I guess I can try the virtual ground, but in that case, I will need a much higher AC input (~30), which would give +/- 15 pre-regulation. The highest (AC) voltage wall wart at mouser is only 24VAC, so that's not going to cut it. I suppose I will have to either a) go for an external PSU that has dual outputs (+/- 15) or b) just get over my fear of high voltage and get a center-tapped transformer. |
|
|
| megajocke |
| I used two half-wave rectifiers to get + and - voltage out of a wall wart. But then my load was small. Just two op-amps in a Linkwitz Transform circuit. It might not work too well if your current draw is unbalanced. I guess using a full bridge and an active virtual ground would be better in that case. |
|
|
| ezkcdude |
Brilliant! That's exactly what I was hoping for. I should be able to use LM317/337 instead of the fixed regs, right? Thanks, mucho.
Edit: Apparently, there is an update to that project (cleverly P05a), specifically, to use adj regs.
http://sound.westhost.com/project05a.htm |
|
|
| paulb |
| There's an error in this schematic; C28 is upside-down. |
|
|
| ezkcdude |
| I've worked up a PCB for a dual supply, which would take an unregulated AC input. This is the first PCB I've ever tried to make, and I'm clearly not knowledgeable about all the "rules", although I've read about many of them. I used ExpressPCB to draw this, and it would be quite cheap ($51 + shipping) to get three boards made with their MiniBoard service. U1 and U2 are LM317/337, respectively. |
|
|
| AndrewT |
Hi,
C3 is connected to common on both pins.
$51 is poor value for three small PCBs. Would stripboard meet your standards?
You could squeeze all the components much closer and then you may get 2 or 3 PCBs onto each board.
Can I suggest a few mods?
1. add an additional pad for each cap to allow for different pin pitch.
2. increase the resistor pin pitch by 0.05inch or 0.1inch.
3. add fuses either to input or to output.
4. output caps footprint seems a bit small. |
|
|
| ezkcdude |
| Andrew, thanks for your comments. I made some changes, if you have a moment to check. As for the stripboard, it seems like a good idea, but I have a few traces on the top layer (green). From what I read, the stripboard only has copper traces on one side of the board. |
|
|
| richie00boy |
Sorry but that layout requires some attention to achieve anything like good performance. You need to take account of component interactions and utilise starring properly.
To save yourself a lot of hassle you might like to visit my website www.readresearch.co.uk and build the 'quality regulated PSU for op-amp circuits' which does exactly what you are wanting and has a good and tested PCB already :) |
|
|
| ezkcdude |
Thanks, richie. Like I said in my original post, I have never designed a PCB before, so I am using this as a learning experience, in addition to being something that will be useful for the DAC I'd like to make. I am hoping that you could be a little more specific about which parts of my design are not ideal? Otherwise, I could obviously just copy your PCB, but that to me doesn't seem very DIY-ish. It would be more like DSE (Do someone else's).
By the way, there is another PCB designed along the same lines as yours, and I gave a link to it on the first page:
http://sound.westhost.com/project05a.htm |
|
|
| richie00boy |
Yes, Rod's design is similar, but mine came out first ;) Mine has room for bigger heatsinks as well.
As for how to design the board well, there is a wealth of information and tips if you study the NatSemi datasheets. |
|
|
| ezkcdude |
| Richie, I was looking at your schematic and I have a couple of questions. 1) What is the purpose of the C1 cap, since the regulators have their own decoupling caps (C6,C7)? 2) What are R1,R2 for? |
|
|
| richie00boy |
C1 snubs common-mode noise.
R1 C4 provide additional filtering, I wanted it to be a very high quality supply. |
|
|
| paulb |
| R1, R2 are part of a CRC pi filter. Very effective for removing ripple. I'd bump up C4 & C5 to 4700 uF. |
|
|
| ezkcdude |
| I've attached an updated version of the pcb. I hope I'm on the right track here. |
|
|
| AndrewT |
Hi,
You have fused between AC and smoothing cap. This needs to be bigger than useful to prevent blowing on start up. either fuse before transformer or after smoothing.
You have only fused one side .
Try rotating one pair of polarised caps and pads so that all polarised face the same way. |
|
|
| ezkcdude |
| Thanks, Andrew. I flipped the lower pair of capacitors. It actually makes the circuit much neater. As for the fuse, I took it out for now. I'm not really sure what you meant by placing one on both "sides". Do you mean somewhere along +/- rails after the filter caps? |
|
|
| AndrewT |
Hi,
move your AC input G to caps common. What about changing the label to AC?
Swap C3 for R3.
You had a fuse in the +ve side of the input, but no fuse in the -ve side to the diode. |
|
|
| ezkcdude |
| Thanks, for catching the bad connection of G (which I renamed). I missed that when I flipped the caps. You're really helping me out here! I think now it's looking pretty good. I've put the fuses back in after the filter caps. Are they o.k. now? |
|
|
| richie00boy |
| Looking a lot better than that first attempt. You could just fuse the primary just fine but it won't hurt to fuse the rails like that. It does depend on what kind of use the PSU will see - if all enclosed with a finished project then IMO not really any need for rail fuses. |
|
|
| ezkcdude |
| Thanks, richie. That's why we come to these forums, after all. If it works, I plan to use it as the PSU for a DAC, and it would be placed in the same enclosure. |
|
|
| ezkcdude |
| This is too much fun! Now, I've moved all the traces to the bottom layer, and I re-sized the filter caps so that they match the specs of the ones I have looked at on digikey (Panasonic, Nichicons). I guess now I can more easily prototype using stripboard, since all traces are on one side. I swear that will be the last change for today. I'm off to the outside to hit some golf balls...Thanks, again, for all the help you guys. |
|
|
| AndrewT |
Hi,
now it's my turn to ask a Q.
Where do you connect the DAC ground to?
Options seem to be G between V+ & V- or to AC2 or to a central ground.
If DAC goes to a central ground, where do you connect the PSU to central ground? from AC2 or from G?
Leave the pads for the two fuses. Link them out if there is no risk of overloading the PSU. Use fuses where you are running power to a more remote (riskier) location.
Similarly keep the CRC pads and link them out when not needed.
You could make alternative holes for the diodes, a bigger pitch and larger diameter to suit 1n540x as well as the present 1n400x. |
|
|
| ezkcdude |
Good question. If I connect G to the DAC ground and then AC2 to the chassis, would that be o.k.? I suppose I could also just combine the PSU and DAC onto a single board, but I kind of like the idea of having a stand alone board, which could be used for other projects and such. It also makes trouble shooting the DAC a little easier, maybe.
Edit: Put the larger diodes in, but also put the option to have smaller N400X. |
|
|
| ezkcdude |
| I accidentally posted the wrong version. Here's the correct one. |
|
|
| AndrewT |
Hi,
yes, multi-purpose board is what I had in mind.
A slight re-arrangement of your routing and you have achieved a single sided PCB which will be much easier to make at home.
I hope someone else comes in to satisfy my curiosity on whether star earthing is needed here, where the ground references are taken from and to. |
|
|
| richie00boy |
| I'm not quite clear on what the star earth question is. The PSU PCB will give you a 0V point which should be the central 0V for the DAC. You would also connect this to the chassis. |
|
|
| AndrewT |
Hi,
the question is whether the PSU could/should be used as an all purpose star ground or the star located off the PSU board.
Secondly which end of the PSU would be better for a ground. I think the transformer tapping before it travels to the PSU board would be better than taking the audio ground from the PCB. But I really am asking not stating.
As for connection to case!
Keep the mains connected case connected to the mains. i.e. apply a safety earth.
Do not connect the audio ground to the mains. |
|
|
| ezkcdude |
| Here's another question for you guys. The DAC should have separate grounds for analog and digital sides, right? So, don't those have to meet up somewhere? |
|
|
| ezkcdude |
So, I only have the N400X diodes, and don't have any 2W resistors. I'm making a digikey order and have added the following to my cart:
1N5401GDICT-ND DIODE GPP 100V 3A DO-201AD ($0.37/10)
WNC10RFECT-ND RES NON-IND WW 10 OHM 2W AXIAL ($0.79/25)
Do these seem like good choices? Also, what about fuses? |
|
|
| zuus |
If you move U1 over to the right slightly you could run a direct track to C4 instead of under R4.
Regards,
Mark |
|
|
| ezkcdude |
| Good point. I'll try that tonight. Any thoughts about my previous question on resistors and diodes? I still haven't found any good discussions of such things (I'm sure there are some out there, so the problem must be with my searching abilities - it seems an important thing these days.) |
|
|
| AndrewT |
Hi,
if you are beginning to get the DIY bug then go buy 100 of each of 1n4004 or 5 and 1n5404 or 5. The discount when buying by the hundred often exceeds 50% and your stock will last for years.
The PIV required for a bridge rectifier is about three times the Vac of the transformer. I suggest you think about the maximum voltage you will require and stock all of the same voltage value.
A 60Vac transformer will need 200V and anything higher (upto 120Vac) will need 400V. Only tube (valve) will need any of the 600V or 800V diodes. There is a slight downside to choosing excessively high voltage diodes, more capacitance, so don't go overboard for just more factor of safety.
2W is about three times 600mW. A standard metal film paralleled three times will give you the power dissipation and be cheaper. The PCB can easily be altered to accomodate three smaller resistors of more accuracy, better temperature stability, and cheaper. Again buying by the hundred is advised. |
|
|
| richie00boy |
| Do you really need 1% tolerance on a simple PSU filter resistor? Just go with whatever suits the board best. If you get into DIY you are going to have to buy a shedload of other components anyway so personally I wouldn't get carried away buying bulk of too much stuff. |
|
|
| ezkcdude |
| Thanks. That seems like a good idea about the resistors. I have some 1% metal film 25 and 49 ohm resistors already, so maybe I can use those. |
|
|
| zuus |
Some things I would do.
Make the track between the two star earths fatter.
Create a little distance between the tracks underneath R5 to reduce capacitance.
Move the track above U1 higher to also reduce capacitance from the track to R3.
Any tracks running in parallel next to each other are going to have a capacitance. On high frequeny boards this is actually used instead of capacitors sometimes.
Regards,
Mark |
|
|
| FastEddy |
I notice that no one here seems to pay too much attention to the ground path on power supply diagrams, photos.
It has long been my habit to put a honking good sized torid coil in the return ground line from the load (amp, whatever) ...
Looking at: http://www.diyaudio.com/forums/atta...tamp=1150979128 ... a healthy coil might go right between the load and the "G" pad on the board.
The idea here being to reduce noise on the return ground path, closer to the load than the main caps. Why? Because this is where the electrons are. This is the usual source for ground loops and this is where RF noise can enter to make an otherwise cool amp run hotter.
No kidding. |
|
|
| AndrewT |
Hi Fasteddy,
can you talk us through that more slowly? |
|
|
| paulb |
| quote: | Originally posted by FastEddy
The idea here being to reduce noise on the return ground path, closer to the load than the main caps. Why? Because this is where the electrons are. |
:scratch: :confused: :drink: |
|
|
| AndrewT |
Hi,
now that your layout is close to what you can use for this project and the next few, can I suggest you squeeze the PCB to give a much more compact arrangement. At a guess you may be able to get all those components into about half the board area. Leave space at the edge (overhanging the edge) for IC sinks. |
|
|
| ezkcdude |
| Mark, I'm interested in your last suggestion about spacing the parallel traces. I guess it is unclear to me when you space traces more closely, and when you should avoid doing that. Maybe I'm thinking of ground loops and inductance vs. stray capacitance. The reason I ask is because if the design goal is to have more space between traces (or maybe avoid parallel traces altogether), it seems contrary to AndrewT's suggestion to make the board even smaller. Care to discuss? I'm away from my computer for a week or so, but I will update the PCB when I get back next week. Until then, I'd love to hear a continuing discussion about these issues. I'm learning a ton from you all. |
|
|
| AndrewT |
Hi,
I don't want to steal the thunder from Zuus.
Simply to point out that shorter traces with the same spacing will have less capacitance. |
|
|
| zuus |
| quote: | Originally posted by AndrewT
Hi,
I don't want to steal the thunder from Zuus.
Simply to point out that shorter traces with the same spacing will have less capacitance. |
Yes, this is true. However if you concider the pcb as a reactance plane then to avoid this, distance and also not having right angle traces (these behave as inductors at high frequency) etc are used to lessen the effects of this (or increase were specifically desired). So by having an overly small pcb the risk would be that all tracks become in close proxomity such that the entire pcb becomes a collection of capacitors and inductors. Also it not a bad idea to keep the diodes away from other parts of the circuit as these can be noisey. So I would not reduce the size because you can, only if you need to. This could be seen as extreme but there can be high currents in power supplies so I think these points are worth a thought.
Regards,
Mark (Zuus) |
|
|
| ezkcdude |
| I'm back with an update. I've made the layout a little more compact, but the overall board size is the same, as it is the mini-board size that ExpressPCB makes. I spaced out some of the tracks and made some tracks fatter. |
|
|
| FastEddy |
Your layout is quite good, well balanced for common mode rejection. Single sided boards = cost effective, reliable power supplies.
I might make two suggestions:
Increase the trace sizes (width) so that a lot more copper is left on the board. (This is more environmentally correct, more cost effective to produce, reduces resistance between components, reduces errors in stuffing and soldering and increases reliability.)
Consider adding an isolated double solder pad close to G (output ground) for the possible installation of a coil or other "hum bucking" component. (I always include a Ferite core coil of several turns in the return to ground path as it changes phasing of the power surges back to ground, improves "turn on" surge resistance and can contribute to even better regulation.
Good show :>) |
|
|
| zuus |
Hi ezkcdude,
Super work!! I do agree with fast eddy though, I think I would still go a little further.
Regards,
Zuus |
|
|
| FastEddy |
AndrewT: "Simply to point out that shorter traces with the same spacing will have less capacitance. ..."
Well AT is certainly right here when it comes to audio signal pathways to and from (I/O) of the signal processing components, etc. ... as stray capacitance and stray inductance all contribute to input and output impedence = of serious concern when DIY project building.
BUT ... when it comes to designing and building power supplies and line filters, stray or extra capacitance can be your best friend ... as extra capacitance on the power and ground lines means extra filtering = more total capacitance = increasing the "noise floor", etc.
DIY Lead Technician on this project, ezkadude: " ... As for the voltages, I want to end up with +/- 12 ... "
Everyone should note that EZKCDude's "isolated ground" design is easily adaptable to any split supply from an AC wall wart or poorly regulated DC wall wart. His layout and design could just as easily output filtered DC of a range of DC voltages from ~= +/- 5 to ~= +/- 24 VDC. The linear regulators, (+) LM317 and (-) LM337, are capable of regulating any input voltage of up to +/- 28 V (peak voltage if unregulated AC). ... any or all by mearly changing a couple of resistor values. The best efficiencies are for split supply output of +/- 9 through +/- 18 VDC ... See:
http://www.national.com/images/pf/LM317/00906301.jpg
http://www.national.com/ds/LM/LM117.pdf ... especially page 8, etc.
...
Notes in passing:
*Finding a decent AC wall wart as a transformer source should not be too difficult, but it may be ... For the +/- 5 VDC through +/- 18 VDC design, the wall wart can have just about any (peak) value of up to 50 or so Volts DC or ~= 50 Volts PEAK voltage AC. ( Example: http://www.halted.com/ccp11916-plug...a000c-80558.htm ... 15 Volts AC is equal to 25.6 Volts peak alternating plus and minus (15 x 1.7) = 51 Volts peak to peak over all ... ~= $3 each ... which should work just fine.)
*Finding a 30 Volt DC wall wart is another matter entirely ... BUT since the EZKCDude's design is patterned after generally accepted "half wave" rectifier designs, the use of an AC wall wart should be acceptable. The above example has a power limit for cool operation of about 200 milliAmps ~= meaning an output limit for cool operation of about 100 milliAmps or 15 to 25 watts overall WITH some healthy heat sinks on the LM317 / LM337 "wings".
*Adding extra solder pads for "Snubbing" capacitors might be a nice touch.
*Adding Ferrite toridal coils to each output (+ & -) AND to return ground (on the left side of the board at V+, V-, G solder pads will "buck" any feedback "hum" from powered components .
*Twisting the output power leads, likewise, will increase "common mode rejection", always a good thing ... one full twist every two or three inches is good.
(UPS did deliver today and my orders for speakers just showed up ... so see ya all after the 4th.) |
|
|
| zuus |
Hi FastEddy,
Capacitance = communication between two points, when this is by design then by applying your rules this will help matters. However one of the main functions of a power supply is to isolate the output from the input noise (i.e. the AC ripple and switching noise of the diodes (and RF)). Thus Capacitance (inductance) in the wrong place could go against this and dirty the ground/output.
I am not disagreeing, in fact I agree (totally actually), but there is a flip side to this statement.
Regards,
Mark |
|
|
| ezkcdude |
O.k. Fattened traces some more. Added a snubber (C9), but is this location o.k.? I was thinking about putting it in the "over" the first star ground in between the electrolytics. As for the ferrite core, do you mean ferrite bead, FastEddy? I have some of these:
http://www.digikey.com/scripts/DkSe...=561546&Site=US
Also, I'm not exactly sure where you mean to put them. After the second star ground, in series with G, for example? |
|
|
| AndrewT |
Hi,
I think you have C9 in the wrong place.
The "snubber" should be in parallel to a diode (two diodes = two snubbers) and the traces should come straight from the diode leads.
The snubber consists of a series combination of resistor and capacitor. You could link out the resistor if you decide not to use it.
The dual holes (0.3 & 0.4inch?) for the smoothing caps need to be relocated slightly. Draw the body outline for each pair of holes and you'll find the cap body fouls the resistor and maybe C9. |
|
|
| ezkcdude |
If you look at the schematic richie00boy posted on his website, the snubber goes all the way across the two diodes and electrolytic caps. That's what I did here. Does anyone else care to chime in their opinion on this?
Because this PSU is really intended to be mainly a "pre-regulator" for a DAC, I'm wondering if it is really necessary, anyway. I could put the "snubber" (whatever form it takes) on the DAC board, itself, right? |
|
|
| ezkcdude |
| Andrew, I added the snubbers you suggested, and moved C9. I also added additional silkscreen body outlines for larger caps. Thanks, for pointing that out. I think the board is now verging on too complex, but if these things help, the additional cost and effort of soldering is not such a big deal. I'm still wondering about the ferrites. Doesn't seem like there is much room left, so I may just leave those out. |
|
|
| FastEddy |
Zuus: " ... Capacitance = communication between two points, when this is by design then by applying your rules this will help matters. However one of the main functions of a power supply is to isolate the output from the input noise (i.e. the AC ripple and switching noise of the diodes (and RF)). Thus Capacitance ([and / or] inductance) in the wrong place could go against this and dirty the ground/output. ..."
Absolutely correct.
Of interest:
a) Several commercially available power supplies use double sided boards ... A few of these makers also leave as much of the copper clad on both sides as possible = extra wide traces, vast "ground planes", etc. These actually contribute to "common mode rejection" of unwanted noise, caused usually from external sources like RF, etc. There is a military spec regarding this in an attempt to reduce effectes of "EMP" and possible electronic interference from the "outside" in the battlefield. These slabs of copper ("planes" or "plates") are effectively parallel with a dialectric between (the non conductive board material) making the board itself a flat capacitor with reasonably good "common mode rejection". (Although EMF (EMP) radiation from certain incidents angles may actually negate this effect, random angles generally do not ... (Military secret = this makes the power supply a little more "stealthy" as well.)
b) Concerning trace widths, usually the fatter the better as in production, errors can creep in so big solder pads and fat traces are the norm in power supplies (but of course not in signal processing). In the case of double sided boards I have seen fat traces and plated through holes used to effectively "twist" conductor pairs over distances of just a few inches, further contributing to "common mode rejection" .... these fat, twisted traces act just like twisted wire bundles in cables ... and the fat traces seem able to accomplish this goal better than the slim ones.
c) Leaving as much copper on the boards as possible also contributes (modestly) to environmental concerns as less ferric-cloride is needed to remove the copper in the bath (a method still used in the Orient and in small batches here in the states) and/or when manufacturing by cutting copper from the boards (by laser or router, etc.), less scrap copper goes into the trash bin ... and reduces "speeds and feeds" in manufacturing time (time = money). This may seem like small details, but when making hundreds of boards it really adds up. .... and the design under discussion here has merit for extensive reproduction = possible market potential = a mass produced wall wart filter for general audio (and military and other) use == $$ = "Anything worth doing, is worth doing for money ..." - Alfred E. Newman (aka William Gates / Mad magazine)
Anyway, these kinds of things are learned by trail and error in the laboratory and shop and usually not discussed in classrooms ... so it is understandable that many here may not know ...
:smash: |
|
|
| FastEddy |
EZKC Dude: " ... If you look at the schematic richie00boy posted on his website, the snubber goes all the way across the two diodes and electrolytic caps. That's what I did here. Does anyone else care to chime in their opinion on this? ..."
Assuming you mean the cap @ C9 ?? ... what material? ... type of capacitor? ... poly? (I may have missed your materials list ... got links?)
You might consider a couple of extra connect points to ground at this location (C9) ... for dual snubbers to ground from the + & - rails = future addons or updates.
;) |
|
|
| richie00boy |
Dude, I placed that capacitor across the whole bridge in that way after careful optimisation. That is the best way to do it. It efficiently snubs the whole bridge.
I appreciate you are not using the full bridge option, but still the single cap should be sufficient to provide any snubbing. In all honesty the noise from such a lower power circuit is difficult to notice/find anyway.
The capacitor should be where it was before. It's useless where you have moved it between the big caps and the way you have run the traces off the big caps instead of the diodes.
As you have noticed it gets messy once you start trying to add a cap+resistor across each diode, and there is absolutely no need to do so.
So stick back with your earlier design (but fix the trace routing) and go make some music :) |
|
|
| ezkcdude |
Well, if the consensus is that I go back to the last version (http://www.diyaudio.com/forums/atta...tamp=1152022977) of the PCB, I will do that. It sure seemed simpler that way. FastEddy, no problem, I will post a schematic (although it is going to be pretty much like richieboy's), PCB, and materials list (which I need to put together) hopefully later tonight, if I get the chance.
By the way, to FastEddy, the first link you posted in your last reply was for a DAC that is my long-term project. This PSU is the first step, but I think it was definitely worth doing, even if it is not the most original design. I have learned quite a bit about PCB design from you and the others (richieboy, AndrewT, and zuus), and tackling the entire DAC at once as a newbie would probably have been insanity on my part. O.k., that's it for now. Gotta get back to my day job. |
|
|
| FastEddy |
Looking good so far:
1) So, is it possible to get a link to a graphic image rather than a .pdf file? This would make it easier for some of us old timers to propose modifications (as some of us can't get it up to edit postscript incapsulated images without buying more software and suffering through the learning curve.)
2) Have you made any boards yet? Will you do so? Would you consider letting some of us have some ($$ pre paid, of course)?
I personally don't want to crank up the old sludge tanks and run off a few myself = too much work, too much mess ... and I might actually have another use for these as prototype for a commercial project ... our catalog at 3dotaudio.com needs some kits for sale, etc. |
|
|
| FastEddy |
http://www.diyaudio.com/forums/show...10&pagenumber=1 ... and check out page 2 ...
This is just one small area where ezkcdude's project has value ... = effects generators & tone control sub-modules for the performers on stage ... electronically, a notoriously noisey environment. |
|
|
| ezkcdude |
| quote: | Originally posted by FastEddy
Looking good so far:
1) So, is it possible to get a link to a graphic image rather than a .pdf file? This would make it easier for some of us old timers to propose modifications (as some of us can't get it up to edit postscript incapsulated images without buying more software and suffering through the learning curve.)
2) Have you made any boards yet? Will you do so? Would you consider letting some of us have some ($$ pre paid, of course)?
I personally don't want to crank up the old sludge tanks and run off a few myself = too much work, too much mess ... and I might actually have another use for these as prototype for a commercial project ... our catalog at 3dotaudio.com needs some kits for sale, etc. |
1->Yes, I can import the PDF into photoshop and export it from there. What format would you like (jpeg, eps, gif?)
2->I haven't made boards. I am planning to get the mini-board service from ExpressPCB, which is $53+shipping for three boards. I could you send you a board (at cost) if you want to try it out. If you'd like to run some tests or something, that would be cool.
The mini-board does not include solder mask or silkscreen. If it works though, I could order the standard boards which do include these (at higher cost, of course). Maybe we could do a group buy if there was enough interest.
As a side note, I noticed you sell the SqueezeBox at that website. I'm a huge fan, and the reason I got into this whole mess is because since I got the SB3, I've been listening to more music than ever in my life. I want to eventually have an entire DIY system built around it. I already have DIY speakers from several years back, so now I'm working on the DAC. When that's finished I'll move onto building an amp. |
|
|
| ezkcdude |
True to my word...
O.k. here's the updated dual supply schematic: |
|
|
| ezkcdude |
I've also put together a sample parts list. These are just suggestions, of course, but these should all have the correct pitch, lead diameters and so forth for the current design.
|
|
|
| ezkcdude |
I forgot to add the regulators to the parts list:
U1 "1.5A, adj reg positive" LM317TNS-ND $1.79 1
U2 "1.5A, adj reg negative" LM337TNS-ND $2.06 1 |
|
|
| richie00boy |
| quote: | Originally posted by FastEddy
I personally don't want to crank up the old sludge tanks and run off a few myself = too much work, too much mess ... and I might actually have another use for these as prototype for a commercial project ... our catalog at 3dotaudio.com needs some kits for sale, etc. |
Fair play to the Dude for coming up with his board and learning a lot along the way, I'm really happy he's achieved that. Questioning why others have done something is a great way to learn.
But you might have noticed he actually copied both my schematic and PCB in pretty much every way, so I find it slightly cheeky you ignore my boards when discussing commercial applications ;) Granted the circuit is quite generic, but still I hope you can see my point :)
Maybe there is some way we can work something out? :) But if you want to get a board of the Dude I can appreciate that, you both being in the same country etc. |
|
|
| ezkcdude |
Richie, I totally agree with you. Without your advice (and that of the others), I wouldn't have the board looking like this. Just take a look at the "evolution" of my design throughout this thread.
That said, this is supposed to be a DIY website. I think it would be unfair for me (or anyone else, really) to profit from the design put forth in this thread, especially since that wasn't my original intention. If there is demand for a group buy, I'd be willing to do that. If people want to use the design and make their own board for themselves, that's fine, too. As for commercial applications, it seems to me that is contrary to the entire concept of this forum. Obviously, I can't prevent anyone from "taking" the design and selling it without my permission (or at least, giving me credit), but I hope that would not happen.
On the other hand...There is an argument to be made that if there is enough demand for this or any other design on the forum, that some nominal fee could be charged to cover the cost of someone's time and effort to produce the board. This would be analogous to the way RedHat profits from supporting and distributing Linux, which is open-source, and supposed to be free. What they sell is their support and packaging of the software. I think we can cross this bridge when we come to it, if necessary.
Richie, just out of curiosity, do you actually sell your board? If so, for how much (in USD) and what is the demand for it? |
|
|
| ezkcdude |
Now, that I'm getting close to getting this board made, I've decided to put the links directly on my personal (non-commercial) website. From now on, all updates (schematics, pcb, pictures, etc) will be posted there.
Schematic:
http://www.cellandtissue.com/ezdac/...y_schematic.pdf
or
http://www.cellandtissue.com/ezdac/...y_schematic.png
PCB:
http://www.cellandtissue.com/ezdac/...upply_board.pdf
or
http://www.cellandtissue.com/ezdac/...upply_board.png
Materials:
http://www.cellandtissue.com/ezdac/..._materials.html
Some issues:
I think the output caps on the regs should be labeled polarized, according to the data sheets. I'm not sure it matters that much, but I thought I'd point it out. Certainly, if polarized caps are used, they need to be oriented properly, so I may modify the schematic and board to reflect this.
The resistor values for the regs have been set to output +/-15 V.
As per FastEddy's request, I posted png files. I tried posting jpegs, but I think my school's server doesn't allow it anymore, for some insane security reasons. |
|
|
| richie00boy |
| PNG is a much better format for schematics and PCBs as it doesn't add nasty blur everywhere because of lossy compression. |
|
|
| FastEddy |
ezkcdude: Ok!! You did quite a lot of work in a very short amount of time. ... and all of it looks very good ... (my advise being worth just about what you are paying for it = $0.00)
A long winded discorse follows ...
Materials list / more suggestions:
1) Rectifier diodes could be 1N4003. Note that 1N4001 is rated 50 VDC, 1N4003 is rated 200 VDC ... FYI: Fairchild diode part numbering scheme is easy to figure out (references below) ... and costs are not important here as all these diodes are about the same $$ unless you are buying hundreds at a time. Other specs are comparable throughout the Fairchild scheme = response time is equal for all series part numbers. All diode manufacturers have equivelents to these Fairchild part numbers. (There is enough down stream regulation that Schottky type diodes need not be considered.)
2) Since it may be possible (in the case of a near lightning strike, strong RF signals, accidental plug into 240 VAC, etc., etc.) for spiking voltages across the input to exceed the expected, nominal 50 Volts DC or AC (peak to peak), it is good practice to rate all components 2.5 to 4 times the nominal voltage at the input and at least 1.5 to 2 times the nominal voltages expected beyond the first stage of internal loads (beyond resistors [or coils] R1, R2). Therefore it is suggested that C1A, C2A and C9 plus R1 and R2 be rated above 100 to 150 Volts.
3) Any and all components down stream to the chip regulators U1 and U2 ... from that first stage of load baring resistance ... should be rated above 100 Volts (DC or AC) including C1B, C2B, F1, F2, C3 & C6. All of the rest should carry ratings above 35 to 50 VDC. (All of this may seem like overkill, but as they say, s**t happens.)
4) It might be possible to use one fuse (or circuit breaker) between the AC / DC input at AC1 instead of dual fuses F1 & F2. (Fuses have their own set of questions, depending on the type used, and can have a small noise contribution of their own, being a type of resistor.) Personally, I would prefer a 1.25 to 1.5 Amp, bat handled circuit breaker ahead of the input at AC1 instead of two 1 Amp fuses == works like a master power on/off switch ... but I would like to see the solder pads and extra traces remain on the PCB.
5) With the exception of C9, all capacitors from AC1 & AC2 through to the chip regulators U1 & U2 can be polarized electrolitic types. (C9 "snubbs" the whole power chain through to the chip regulator inputs and C5 and C8 can "snub" the outputs.)
6) Likewise, any resistors from AC1 & AC2 to the chip regulator inputs of U1 & U2 can be of any type, carbon or whatever ... but resistors R3, R4, R5 and R6 should be of a low noise type like metal film or other "high tech" construction ... as there is a potential for a noise contribution from carbon resistors ... but you already knew this as per the materials list.
7) And of course C4, C5, C7 & C8 should be plastic "poly" or Tantalum ... but you already knew this as indicated in your material lists.
8) Power values for any resistors (or coils [inductors]) in the power chain should be compatible with the "worst case" requirement. In this case R1 & R2 should be 5 Watts or more, combined and of small resistance value ... or use coils instead to "load" the capcitors ahead of the chip regulators. Power values for the voltage reference network, R3, R4, R5 & R6 can be 1/4 Watt or less as no significant current should pass here as long as the R3 & R4 combined are >>= 1000 ohms (~=1.2k, combined) ... as you indicate in the material list.
9) Although National Semi insists that the chip regulators (LM317) can handle up to 1500 milliAmps, this would require very large heat sinks and in any case should be considered as a "worst case" scenario, National apparently believing that "infinite" heat sinks are manufacturable. Us mortals should consider that the LM317 should easily handle 500 milliAmps each, with a generous heat sink. (Note that by using LM338, a more raesonably sized heat sink will do nicely.) Note that any heat sink on these chip regulators should not be connected to any ground point. ...
10) If fixed regulators, like -7815 & 7915 types, were used, a) the heat sinks should be tied to ground (the ground plane, board copper acting as added heat sink) and b) the whole network R3, R4, R5 & R6 & C4 & C7 could be removed. (!) ... But I would not remove any of the solder pads and associated traces = making the boards usable for either type of chip regulator. (See references to 7805 through 7818 and 7905 thru 7918 below =all available for about $5 per ... or much less in qty.)
11) Ferrite Beads may be placed close to the input legs of the chip regulators U1 & U2 if the need seems great ... unless coils are used instead of R1 & R2 resistor. This can contribute to chip regulator effesencies = less heat and better regulation, less noise, etc. ...
The circuit / more suggestions:
Generally, the circuit is fine as you present it and follows generally accepted practices for half wave and full wave rectifying power supplies. You are not actually reinventing any new wheels here, except for making accomodations to the stricter requirements of quality, high end audio. ... And with the boards are fully stuffed with components, the finished device can be used to filter unregulated AC wall warts or transformers OR poorly regulated DC wall warts.
Because of the care taken in component selection, the overall efficiencies of this linear power filter /supply should be quite good.
Provided that the resistors R4 & R6 are carefully selected (assuming R3 & R5 are fixed values), input voltages could range from 12 Volts (7.5 VAC (~= 12 V peak to peak or 12 VDC) to 30 VAC (~= 50 V peak to peak or 50 VDC) across AC1, AC2 and produce well filtered, well regulated, split outputs of +/- 5 VDC to +/- 18 VDC ... !
(I don't recommend using variable resistor at R4 or R6 as all adjustable resistors have the potential of making a noise contribution. ... but you already have discovered this.)
References:
http://en.wikipedia.org/wiki/Schottky_diode
http://www.fairchildsemi.com/ds/1N/1N4004.pdf
http://www.national.com/pf/LM/LM317.html & http://www.national.com/pf/LM/LM338.html
http://en.wikipedia.org/wiki/Ferrite_bead & http://en.wikipedia.org/wiki/Inductor
http://www.fairchildsemi.com/pf/LM/LM7815.html & http://www.fairchildsemi.com/pf/LM/LM7905.html (7915 info)
More on the PCB later ... if you can stand these long winded articles. |
|
|
| ezkcdude |
Thanks. I have definitely "grown up" during this thread.
Moving on...I think we need to be careful here not to overspecify. With higher voltage and current ratings comes more expensive and more importantly, larger components size. At some point, the resistors and caps (probably the major concern) will be physically too big. FastEddy, you can print out the png, it is drawn exactly to scale. If you already have a supply of some of these parts, can you just see if they fit the current board, in terms of pitch, lead diameters, body outlines, etc? That would be helpful. I'm sure the board will fit the parts I have listed, but I don't have much else on hand to try out, and I don't really have the time to sift through the entire digi-key catalog. I think the board can accomodate fairly beefy components, but obviously there are limits.
If people have recommendations for specific components, please give me a part number (digikey, mouser, newarkinone), and I will look it up and see if the specs fit the current board design. No problem.
Edit: I did some quick checking. With the caps, if we go to 50V rating, then we should lower the capacitance to 3300uF, and if we use 100V, then the capacitance is 1000uF for the same body outline size. This is just to give an idea. It looks like we may be safe going to the higher voltages, but may have to give up a little in terms of capacitance. |
|
|
| FastEddy |
More notes on material list options.
from: http://www.cellandtissue.com/ezdac/..._materials.html
D1, D2 50V-1A 1N4001GDICT-ND ....... suggest using 1N4003 = 200 V.
C1,C2 4700 uF, 25V 565-1068-ND ....... suggest using higher voltage 35 or 50 V.
(also the need for this much uF may be overkill, 1000 uF should be plenty)
R1,R2 33R metal film, 1/4W, 1% P33.0CACT-ND ... metal film may not be required, 5 watts suggested.
F1,F2 125V, 1A F868-ND ...... or use single breaker at AC1 input.
R3,R5 100R metal film, 1/4W, 1% P100CACT-ND ..... N/C
R4,R6 1.1K metal film, 1/4W, 1% P1.10KCACT-ND ... N/C
C3,C6 10uF tant, 25V, 10% 399-3565-ND .... or 50 V., 10 to 20 uF Plastic Poly Cap.
C5,C8 10uF tant, 16V, 10% 399-3563-ND ..... or 50 V., 20 to 50 uF Plastic Ploy Cap.
C4,C7 1uF tant, 25V, 10% 399-3528-ND ... N/C or 35 V.
C9 100nF metal polyester, 250V, 10% P10967-ND ... actually as a snubbing cap, this value may be high. (small value metal poly caps paralleled with large electrolitic or Tant caps work better at higher freq filtering.) ... but N/C is fine.
U1 1.5A adj pos reg LM317TNS-ND ... N/C
U2 1.5A adj neg reg LM337TNS-ND ... N/C
Heat sinks ??
........
See the previous notes about using Fairchild 7805 / 7905, 7809 / 7909, 7812 / 7912, 7815 / 7915 or 7818 / 7918 as fixed voltage regulators.
Also added references:
The Op Amp Cookbook by Don Lancaster (circa 1975) and more recent books about using adjustable and fixed voltage chip regulators. FYI: all of these linear chip regulator designs are derived from original work done with the (very old) uA741 op amp from Fairchild and National. |
|
|
| ezkcdude |
| Can you give me some part numbers here? I doubt any 5W resistor will fit in the space I have provided on the board. If you can be more specific about components (part numbers, please), then we can go from there. Otherwise, it's just too much work for me. |
|
|
| FastEddy |
ezkcdude: " ... I think we need to be careful here not to overspecify. With higher voltage and current ratings comes more expensive and more importantly, larger components size. At some point, the resistors and caps (probably the major concern) will be physically too big. ..."
Yes you are correct about the sizes as voltage rating goes up.
Note that you may be using to much capacity in C1A, C1B, C2A & C2B ... so by reducing the size (to maybe 500 uF or 1000 uF) the voltage rating can be increased without increasing physical size too much. Considering that the (total) voltage potential might exceed 250 volts In the case of a short circuit from European power outlet ... oops, get the fire extinguisher. For Kits and DIY, we may not worry too much, but a lawyer might get a hold of one of these ... Is 100 volts too much to ask? ... how about 50 volts. |
|
|
| FastEddy |
ezkcdude: " ... the png, it is drawn exactly to scale. If you already have a supply of some of these parts, can you just see if they fit the current board, in terms of pitch, lead diameters, body outlines, etc? ..."
It is quite common practice to have very large capacitors mounted "off board" and connect them to the PCB with (in this case) #16 to #18 AWG wire, soldered to taste directly to the existing solder pads on your PCB design without changes. |
|
|
| ezkcdude |
I agree. I think 4700 was overkill. I suppose I won't know until I actually test the board. Like I said before, for the caps, the part sizes at 100V will probably be fine, if we lower the capacitance. I think I should just get the board made, and try out several different smoothing caps. With my scope (Tek 465B), I should be able to make out the amount of ripple, but I think it will be hard for me to "see" the differences that the components elsewhere have on high frequency noise. Am I right?
As for the offboard caps, I've seen that. Personally, I'm not too keen on it, because I think you have to worry about the mechanical connection, but obviously, it's been done. Well, I guess what you're saying then, is that if people want to mod the board with gigantic caps, it could be done regardless of the final board size. That's a good way to think about it, I suppose. |
|
|
| FastEddy |
richie00boy" " ... copied both my schematic and PCB in pretty much every way, so I find it slightly cheeky you ignore my boards when discussing commercial applications Granted the circuit is quite generic, but still I hope you can see my point ..."
Oh, my ... Sorry about that. I have not paid too much attention to all the different references to your circuit designs.
Please understand that your work is not under appreciated ... http://www.diyaudio.com/forums/atta...tamp=1152158847 ... etc., etc. is probably exactly the same as your workup ... so even with changes to values and specs of components, your pictorials / circuits / boards are probably worthy of mention at least ... I'll be the first to stand up and cheer = Thanks for all the work.
If you want Koo Doos when it comes time to publish, print, post the html code, etc. et al ... we will certainly at least mention your previous work ... and if we ever make more than a few bucks on this (unlikely) then I owe you dinner at a CES show ... (I have to be careful here, as knowing those guys at National, they may want drinks around as well!)
My very first job as an electronics technician was to build a better power supply than the bosses ... I failed and was fired (a good thing it turns out) ... so I made it my business to learn about AC to DC and DC to DC conversion, power supply design ... :smash: |
|
|
| FastEddy |
ezkcdude: " ... With my scope (Tek 465B), I should be able to make out the amount of ripple, but I think it will be hard for me to "see" the differences that the components elsewhere have on high frequency noise. ..."
Yes = correct. Ripples on the 'scope can occure all over the place ... that's why I keep mentioning coils and inductors as needed ...
Pure sign wave /\/ ripples become \/\ when passing through a coil and depending on the ripple frequencies v. reactance (& reluctance [in Henrys]) can actually be made to cancel out ... so everything down stream works less hard = increases in efficiencies = better noise rejection.
"All of the worlds problems can be resolved by fixng the impedence mismatch ..." - Bob Porter / Mad Scientist. |
|
|
| FastEddy |
ezkcdude: " ... I did some quick checking. With the caps, if we go to 50V rating, then we should lower the capacitance to 3300uF, and if we use 100V, then the capacitance is 1000uF for the same body outline size. This is just to give an idea. It looks like we may be safe going to the higher voltages, but may have to give up a little in terms of capacitance. ..."
It could also be possible to decrease C1A, C1B, C2A, C2B and increase C5 and C8 accordingly and produce almost identical results on your 'scope.
Overall filter capacity is not actually "determined" by the chip regulators, but increasing capacitance at the output makes it easier for the chip regulators to do their work.
... and if coils are used in place of or in addition to R1 & R2, then capacitor sizes across the bridge can be reduced further ... (these coil inductor work exactly like the coils in a crossover network = the L/C time constant increases with L and decreases with C for a given frequency, be it a noise frequency or an audio frequency at a "roll off" knee.) |
|
|
| ezkcdude |
I think my earlier concerns about cap sizes may have been a little too hysterical. The body outline currently drawn will accept the following Nippon SMG caps (and I'm assuming it will be similar with other brands):
25 V - 12,000 uF
35 V - 6,800
50 V - 4,700
100 V - 1200
That should be plenty capacitance, even if one chooses to go up to the 100 V rating. I think I will go with the 50 V rating, but maybe use 2200 or 3300 uF. FastEddy, I guess I'm about ready to order the board. If your interested in getting one for yourself, send me a PM. |
|
|
| ezkcdude |
| So, would it be possible to simply replace one or more of the resistors (R1 or R2) with a coil without changing the design? You can have resistors and inductors in parallel there? If so, that would be cool. And what value inductor would be appropriate? |
|
|
| FastEddy |
ezkcdude: " ... I guess I'm about ready to order the board. If your interested in getting one for yourself, send me a PM. ..."
I have not had time to thoroughly review the PCB layout ... but from what I remember = no problems. |
|
|
| FastEddy |
ezkcdude: " ... would it be possible to simply replace one or more of the resistors (R1 or R2) with a coil without changing the design? ..."
Actually, yes ... keeping in mind that paralleling a resistor and a coil lets AC through to the limits of the L/C time constant, but a coil lets DC through without significant resistance, so that the resistance of the paralleled resistor is negated = not applicable. In otherwosrd, coil inductors of healthy wire sizes don't restrict DC power at all, but resistors do ... so having a resistro R1 and a coil at R1 = almost no resistance at all to DC = arguably a good thing, except for the "wasted" resistor.
It gets complicated from here on :hot: ... suffice to say that a couple of coils of different values substituted at R1 and R2 actually produce a broad range of frequency filtering and "hum bucking" feedback reactance ... to further complicate this ... the "black art" trick is to use an additional coil on the ground bus between the load and G on the PCB. This then allows further reduction in capacitor capacity (size) , thus allowing an increase in capacitor voltage rating ... tweaking and tweaking until capacitor sizes become cheap and PBC real estate becomes managable. :smash:
References: Maxwell's Equations and anything Nikola Tesla ever wrote on electromagnetic induction, then consult the ARRL Handbook (any issue from 1965 on) ... :smash: :att'n: |
|
|
| richie00boy |
There would be no point using 100V caps as you would never put anything like that much voltage on the board. For 100% reliable operation the max voltage each regulator input should see is about 40 volts.
I'm not trying to clamour for attention :) But it would be nice and appreciated if you could provide links to my website somewhere. www.readresearch.co.uk
Good luck with your projects, both of you. |
|
|
| FastEddy |
richie00boy: " ... There would be no point using 100V caps as you would never put anything like that much voltage on the board. For 100% reliable operation the max voltage each regulator input should see is about 40 volts. ... "
Yes, absolutely ccorrect. Using a 30 VAC wall wart or transformer, the peak to peak voltage is about 52 volts ... but if someone accedentally plugs that 120 VAC primary / 30 VAC secondary into an outlet in Europe ... oops, grab that fire extstingisher ... secondary voltage goes to 104 VAC or more :hot:
Anyway, there are always opportunities for a screwup like me to do the wrong thing and cook the electronic goose. |
|
|
| ezkcdude |
| quote: |
I'm not trying to clamour for attention :) But it would be nice and appreciated if you could provide links to my website somewhere. www.readresearch.co.uk |
I certainly will, once I have actually built the thing. I think I will dedicate a page on my website to my DIY projects. That ought to bring you at least 2-3 visitors/year ;). |
|
|
| ezkcdude |
O.k., did some more heavy lifting...
See post #74 for links:
http://www.diyaudio.com/forums/show...5536#post955536
1. Fattened traces near regs
2. Changed body outlines of smoothing caps, regulator caps, and diodes to be more uniform and to better match materials list.
3. Spaced the wire connectors (10 mm) and re-sized them (18 AWG) so that terminal blocks can potentially be used. I need to order some, and check this.
4. Added higher quality (higher rated) components to materials list (in italics)
Now, we're getting down to brass tacks. I'm going to order up some parts, make sure they fit, and then once that's done, order some boards. |
|
|
| FastEddy |
see http://3dotaudio.com/doodles.html ... a doodled up version of ezkcdude's PCB layout with more copper left behind.
The idea would be to save expenses (costs) in large production, but probably is not needed for short runs. (Speeds & feeds in the tank or on the mill be about 1/2)
:smash: |
|
|
| ezkcdude |
| Yeah, I thought about filling the board like that. Is it a good idea? My understanding is that it could lead to problems with the board warping. Maybe copper could be left on the component side to balance this out? Would that increase or decrease noise? I'm happy to make these changes. Also, I'm going to fiddle around with adding some protection diodes for the regs. There may be room for some N4001's. |
|
|
| FastEddy |
I suppose it could if it gets hot enough ... but I have not seen it happen normally.
Best reason for leaving copper on the board is costs savings in manufacturing = reduced speeds and feeds ... second best is environmental concerns = less copper & plastic in the trash ... but it is possible do do really neat things with the extra copper real estate ... like write the component part numbers in the copper, etched or milled ... R1A R1B R1C here > ... and thus eliminate the silk screening altogether >>> :D
And another thing: to give an idea of how effective the extra copper is in shielding ... note that virtually all WiFi and GPS antennas in handheld devices are now done in copper on the PCB board = same as with RFID tags ... and if antennas work, so do well designed shields.;) :smash: :cool: :att'n: |
|
|
| ezkcdude |
| O.k. but I don't think you answered my question. Should I leave copper planes on the component side? |
|
|
| ezkcdude |
So, here I filled the entire bottom layer with the copper ground plane. Will this actually perform better? If so, it seems a no-brainer to keep it.
Also, I've added the protection diodes (D3-D6), which took some jostling about. |
|
|
| richie00boy |
In this instance a copper filled area (normally you would connect it to 0V to gain a ground plane) would not hurt. In sensitive circuits it can increase capacitance to ground which can result in problems, but you should be fine here.
As has been said it's a more environmentally friendly way to do it, which I like :) |
|
|
|
