xerxes
Been here a while!
Posts: 1,115
|
Post by xerxes on Jul 4, 2006 2:14:31 GMT
I was wondering about using battery power for multiple output voltages. I saw this example circuit in the LM337 adjustable voltage regulatator datasheet: From which I came up with this: Basically I've added a bunch of large value radial electrolytic capacitors, to lower the output impedance as per Rick's advice for the headphone amp battery PSU. Although I may use more lower value capacitors than I've shown in the diagram. Questions:1. Am I correct in assuming that the 0.1uF and 1uF capacitors would be non polar "box type" capacitors and not radial electrolytics as shown in the first diagram? 2. As this is for DC, I don't need a bridge rectifier doohickey, but how do I get a 0v/ground for a +/- 15v output using a DC power source? What do I do with the third connection: 3. Could I connect a whole bunch of regulators to a single 24v DC source in order to give a variety of different outputs, like this:
|
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Jul 4, 2006 2:30:06 GMT
|
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Jul 4, 2006 2:45:54 GMT
|
|
xerxes
Been here a while!
Posts: 1,115
|
Post by xerxes on Jul 4, 2006 2:55:57 GMT
I've got no idea "how much is enough", so I thought better too much than not enough. ;D I thoght you were the scary monster?! Also, I need to find out how much power the DAC uses all up. If it's fairly modest, 1 amp or thereabouts, batteries will be quite feasible. However, if it's quite high it means big ass batteries to get a decent running time and it all starts to get expensive. However, I did consider that I could use a suitable DC power supply, perhaps something like this as a backup and for times when I have music on in the background. In which case the caps would further smooth the DC input from the mains PSU, wouldn't they?
|
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Jul 4, 2006 3:24:04 GMT
Part-2 :
You are right to want a regulated battery output voltage.In plain terms (from Maxim app note 737)
Why Power-Supply ICs?
The chief purpose of most power-supply ICs is to regulate. These devices take an unregulated input voltage and provide a regulated output voltage, that is, an output voltage that remains steady despite varying input voltage or output current. This accounts for the names linear regulator and switching regulator. The exception is the charge pump: Depending on the specific device, a charge pump's output can be either regulated or unregulated.
Sometimes regulators create a regulated output voltage from a regulated input voltage. In that case, the regulator's function is to change the input voltage to another voltage level, without necessarily improving the voltage regulation.
You might be tempted to power a circuit without a regulator, and in fact in some cases you could get away with this. You could, for example, power portable equipment directly from a battery. But this approach usually leads to problems. The circuitry within portable equipment most likely operates correctly only within a certain narrow voltage range. This is especially true with microprocessors and memory, particularly if high speed is needed. For microprocessors, memory, and many other types of circuitry, the voltage range over which the battery operates could extend beyond acceptable levels. Adding a regulator ensures that your circuitry receives the appropriate voltage.
The battery's internal resistance could also present a problem if a regulator isn't used. This difficulty arises because circuitry within portable equipment often demands a varying level of supply current. This varying current, when drawn from the battery, creates a varying battery voltage due to the battery's internal resistance. Portions of the circuitry might object to these battery-voltage variations (in other words, the power-supply rejection ratio of the circuitry's various components might not be sufficient to reject these voltage variations). To combat this problem, a regulator maintains a steady output voltage despite these varying load currents. A regulator provides this steady voltage, because its active circuitry maintains an output resistance that is significantly lower than the battery's series resistance
that last part about the varying voltage of the battery is real important when powering the analog section which is a dynamic drain,the first part for the digital section where dead on voltages are a must.
|
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Jul 4, 2006 3:25:09 GMT
|
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Jul 4, 2006 3:26:43 GMT
and oh yeah,you can lose the 0.1uf cap on the battery pack output (doing nothing but being there) but do put one across every single regulator IC input
|
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Jul 4, 2006 3:30:07 GMT
quick question-why the four 3.3 v regulators and no 5.0 volts shown ? Typo ?
and do you think dual mono bipolar 15 volts for the analog section needed here ?
I would suggest you simplify down to a single 3.3 vdc buss,a single 5vdc buss (or do you need +/- 5VDC ?) and a single +/-15 VDC buss then decouple at the chips.
|
|
xerxes
Been here a while!
Posts: 1,115
|
Post by xerxes on Jul 4, 2006 15:10:42 GMT
There is no 5v, because the 5v was for the optical link on the DAC. I don't need an optical link, so I'll leave it off. I'm following the layout, with regard to separate voltage regulator boards, of the original design. The original design used separate VR boards for the left and right channel digital and analogue sections of the DAC. The design uses 2 separate PCM1794 DACs in dual mono configuration. I assumed that keeping everything separate all the way back to the power source, be it mains or battery, would in theory give some performance benefit; reduced crosstalk? To some extent, this is not a "how cheap can I make it", but a "how good can I make it" exercise. If I make the VR boards myself using stripboard, point to point, or possibly even use an etch resist pen to draw the circuits, the most expensive parts will be the large value electrolytic caps, everything else is pennies. Each VR board might cost around £10, so I'm not worried about corner cutting to save £20 or £30 on a project which may cost £400 plus in total. Given your comments about the "slightly excessive" use of capacitors in the previous diagram and the battery PSU article you linked to, does this look better: And have I understood the +v 0v -v thing correctly:
|
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Jul 4, 2006 15:44:56 GMT
Ad copy selling features and common sense are usually miles separated. What no toslink for your portables ? looking at your "new" schemo it is a 100% improvement over the former "we gonna start a war beginning with this bomb right here" power supply but I would personally still change a couple of things. I got some other pages open at the moment trying to work up a USB CODEC but will wrap up soon and then check the actual DAC data sheet and the actual product sheet to see what would be my preference (and why) then you can choose the best path for YOU from there. I think we can bring this all the way down to a single 24VDC battery pack with a single 3.3 VDC supply buss and a single bipolar +/-10 VDC for the analog section which should be fine while keeping the cost to build WAY lower. The ultimate would be to use a single +5VDC regulator then individual TL431 shunt regs with small decoule networks right at each power oin of the digital I.C.'s but board space says that may not be possible.I will also check to see if you even NEED the output stage hence output stage power. I have found with damn near every single one of the DACs I have ever made the output section can usually be entirely eliminated and still have enough voltage drive for full scale DAC drive so i need to check the architecture of this product to see what the deal is on the output then check it against the data sheets for a maybe simple answer to the power dilemma
|
|
xerxes
Been here a while!
Posts: 1,115
|
Post by xerxes on Jul 4, 2006 16:19:46 GMT
What portables? The DAC as I build it would have the following inputs: SPDIF via a standard RCA input USB AES/EBU via an XLR socket, not that I have anything to use this with. There are also output headers on the board for the following: L+R Balanced output via XLR 2 x L+R Unbalanced outputs via RCA sockets I have no use for balanced outputs at the moment, but if I change my amp, I may well have in the future. Two sets of RCA outputs will be useful. One set for my main amp and another for my headphone amp. Also, can I do this:
|
|
Stormy
100+
Advocates putting smokers in a "Sin Bin"
Needs to learn to keep his big mouth shut.
Posts: 153
|
Post by Stormy on Jul 4, 2006 17:54:38 GMT
Also, can I do this: Forgive me if I'm being stupid (and please don't think that I'm being patronizing, because i really am not trying to be), but as far as I know it's very difficult to produce +/-24V with two 12V DC batteries. From what I remember of my electronics, I think you can only do this: or this: It works fine with AC though.
|
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Jul 4, 2006 18:24:03 GMT
Not good. You never want to run two loads from the same output or you get contamination of the signal to both.I would think if you are worried enough about a dual mono supply that has questionable merit you would REALLY worry about this type of connection where both amp input loads will be connected together at all times so no matter which one you are listening to the other is robbing drive power from the DAC Unless of course there are separate amp stages for the SE and BAL stages then disregard it is.I assumed he would be using all positive regultors (at least I am assuming this ) with the "-" output of V=REG 1 tied to the "+" output of V-REG 2 so becomes the "common" point or analog ground otherwise you are right,Two 24 volt batteries would be needed with the common at the center
|
|
xerxes
Been here a while!
Posts: 1,115
|
Post by xerxes on Jul 4, 2006 19:51:49 GMT
OK, I think (hope) I've got it: Yes? ....or No? If it's right I can go back to using one 24v battery for all 6 regulator baords: Alternatively, I could do this, but presumably that would drain the batteries unevenly? I'm not certain, but looling at the board, it looks like there are 4 Burr Brown op-amps for each of the left and right output sections, to the right of the board in the picture: I will confirm, but I'm guessing that each output does have a separate amp stage.
|
|
xerxes
Been here a while!
Posts: 1,115
|
Post by xerxes on Jul 5, 2006 13:35:22 GMT
I came accross this article in a post on another forum: www.wenzel.com/documents/finesse.htmlSome of it goes over my head, but I get the general jist of it and it looks as though it would even have a benefit with battery power.
|
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Jul 5, 2006 15:24:23 GMT
nice article,one I have printed to hard copy and added to my files for a while now but long fogotten I was looking at the EVM-PCM2704 pdf focus.ti.com/general/docs/lit/getliterature.tsp?literatureNumber=sleu049&fileType=pdfand noticed they were using the REG1117-3.3 LDO to get the on board 3.3 VDC from an external 5VDC power source so i downloaded the pdf for the reg chip focus.ti.com/lit/ds/symlink/reg1117-33.pdffocus.ti.com/docs/prod/folders/print/reg1117-33.htmlPerfect ! Exactly what I needed in a previously overlooked part (something to soon change ) I figure my final USB interface whatever shape it takes will use this as the "on card" V-reg while using my current "Bipolar 5VDC Box" as the external power source elimination any possibility of delays in the turn on sequencing when the DITB and USB DAC are used as a pair (if the final DAC is the 2704,likely). you may want to look this part powered by an LM7805 (I am using an LM350K set for five volts in mine ) +5VDC pre-regulator as well for your needs as well considering it is a high current output LDO which relaxes the parts count if you needed to add a pass transistor to get that kind of current output. That would leave only the analog section PS to deal with BTW-have you actually purchsed the kit yet and if so what are the opamps ? Maybe +/- 15 VDC is not an essential making the demands on the battery section easier to deal with if you have NOT already purchsed the kit then you need to look at the K&K DAC that uses the same chip but has the 3.3 V regulator already on board and by having NO output stage on the DAC board but many options for one is about as flexable as it gets for the chipset.Add to that the DAC support/description/design detail pages at Raleigh Audio and it may be right for you if you have not already ulled the trigger I personally have never heard a 1794 based DAC but have not once read anything but good reports on the sonics so ...........
|
|
xerxes
Been here a while!
Posts: 1,115
|
Post by xerxes on Jul 5, 2006 15:53:49 GMT
I haven't actually purchased the kit yet, I'm still mulling it over and working out what the "all up" cost will be. There are also some issues I wanted to get clear in my mind before I go ahead; obviously I want to end up with a working DAC at the end of it. There are a number of things that attracted me to this particular kit. I wanted at least a USB and a SPDIF input. I like the option to use balanced output, which is favoured by many amp makers and is often said to sound better than the RCA input with amps that have both. Lastly the PCM1794 DAC chips; like you I've seen a few positive reports on this DAC and they do seem to find their way into a lot of expensive kit, so they must do something fairly right. By the way, have I got my VR circuit right yet?
|
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Jul 5, 2006 18:48:58 GMT
good decision makes sense these days with the transition from stand alone players to "everything in the computer box" going on.That we may will see the end of most home media players,one for each format, except at the ultra high end I have little doubt.My feeling is convergence is late but still on its way with windows Media Edition the first real shot if it ever takes hold and that means USB over SPDIF even though Fire Wire may have been the better way. Off hand (and without my lazy butt having to go back and read the page ) do you know what type of USB to DAC chip interface is being used ? If I2S from a PCM2706 probably your DAC of choice the better way but if a PCM2704 to spdif then on to the spdif receiver of the DAC you may just want to pass go and collect $200 now by going with this pcb for $39 www.hagtech.com/hagusb.html mated to this fully built DAC board for $249 www.kandkaudio.com/digitalaudio.htmlthen add a hefty five volt power supply to power the whole deal,battery or line,then your output stage Is the current "in" thing and many have balanced just to do it and for no other reason.Ask most a simple "why" and they usually give some vague "well....um...er....um....someone said it sounds better....um......somwhere.......er....um...once" Not knocking it but if it is ad copy or me too-ism making you think you need balanced outputs you should re-evaluate need. In this particular context not as much of an issue with the PCM1794 by nature balanced due to the +/- DAC outputs but in other areas a real issue that has audible consequences,especially if any stages need single ended -to-balanced or balanced-to-single ended conversion which means additional stages you would noty have otherwise needed-not good at worst and not cheap at best. maybe that is why I like the Rakk Dac idea so much-choices.In the DAC you show more than likely the XLR balanced outs each use a single opamp stage,one for each polarity (the -out and +out) which is then followed by a differential input opamp or INA chip for the BAL to SE conversion.Not positive on this but is likely. That means two opamp stages between the DAC output (if true) and the RCA jack minimum. If you look at fig.23 from the data sheet ; focus.ti.com/lit/ds/symlink/pcm1794.pdfit shows the differential IV stage but NOT the analog filter sections or BAL to SE converter sections other than as a "box" . BTW-I am also seeing +5 VDC listed as the analog section power to the DAC chip so it looks like you are back to +5V DAC Analog,+3.3V DAC Digital and then whatever your output stage needs.Things are never as simple as first glance Maybe yes,maybe no. A good DAC chip does not mean you will have a good DAC just just that you MAY in right design.If it was all about the part then even budget commercial CD/DVD players using the 1794 chip-many do- would kick ass without any changes but we know that is not true Nope ;D you can't use a positive regulator and a negative regulator on a single polarity battery of +24 volts to get a bipolar 15 volts.you need to either increase the battery voltage to the combined output of the two plus voltage drop of the regulator (plus a bit of headroom for battery disharge sags) .Plus your ground (actually now a "common" or "center voltage reference" there being no ground anywhere ) is all wrong. 1-place the caps following the battery directly across it.you are not trying to filter the regulators but drop the battery impedance.Add single big low ESR cap from battery "+" to battery "-" then a 10uf/0.1 uf pair at the input to all V-Regs and you will be fine.Consider it like a distrubuted power system with the battery tasking the place of the normal AC transformer and rectifier-a simple clean DC source 2-you need a 5 VDC DAC supply,check the sheet for current draw 3-you need a 3.3 VDC DAC supply unless it is already on the pcb,it should be,in which case the single 5 VDC will take care of everything but if not needs to be followed by a 3,3VDC regulator.No resistor network is accurate enough under load and in real use to be of any use 4-you need to generate a negative 15 voltage from the poistive 15 voltage using a single 24 VDC source. A current pump comes to mind if you must have a split 15 volt supply but can be mirror image regulators (one positive,one negative) if you can live with +/- 10VDC supplies. Probably other ways but none come to mind right now that are regulated
|
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Jul 5, 2006 19:20:33 GMT
|
|
xerxes
Been here a while!
Posts: 1,115
|
Post by xerxes on Jul 5, 2006 21:03:53 GMT
It looks like the total power required for the DAC is around 6 amps, which means a couple of 60 Ah batteries for a 10 hour run time. In addition I would need a fairly hefty 8 - 10 amp 24v charger for for a 7 to 8 hour recharge time. All in all, it's not looking great for battery power. On the upside, I don't need to worry about trying to create a +/- regulator from a battery source, which I don't seem to be getting anywhere with. ;D
|
|
xerxes
Been here a while!
Posts: 1,115
|
Post by xerxes on Jul 5, 2006 21:40:17 GMT
I've just been looking at the system block diagram and the DAC uses a PCM2706 with the USB input, so I'm guessing it doesn't convert to SPDIF, but I'll check. UPDATE According to the designer: I think we may also see the "voltage finessing" circuit as an add on board that can be used to further improve the output from the existing voltage regulator boards.
|
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Jul 6, 2006 2:03:17 GMT
USB>2704>SPDIF Receiver>I2S>I2S Interface>DAC Chip is no more than the same as using the HagUSB (@ $39 for the pcb) to get a USB input to the SPDIF Receiver !
It almost looks like the DAC you reference is attempting to get every known feature possible in to a single DAC board in the same way some folks try to use whatever parts are "said" to be the best or currently popular-because individually someone said they were the best.
If I have learned anything about audio gear over the years is you can use combinations of "the best" of everything and end up with total shit sound or you can have a plan using what look to be comprimises on paper (or the ad copy) yet in unison transcend the parts in SQ because you got the matchups right
Good design with a plan trumps features or catch phrases so know what you need and why you need it before buying in.You may find your requirements can be met for less once you know what what the end goal is.
|
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Jul 6, 2006 4:47:42 GMT
|
|
xerxes
Been here a while!
Posts: 1,115
|
Post by xerxes on Jul 6, 2006 13:32:45 GMT
Nope, at least I don't think so. Here it is: www.audiodiylab.com/modules/news/At the moment it's still a work in progress, but I've been following progress in the forum and asking the odd stoopid question. It has all the features I want, but of course there's the big question; "how does it sound". For the answer to that I will hold off being the first builder and wait to see some feedback, but ultimately there is only one way to really find out. I hear what you're saying; my first "proper hi-fi" set-up was a collection of "five star, best buy" components and to be honest the end result was rather ordinary. I currently have a one make system, with the exception of the speakers, which are a different make, but are a popular partner for my electronics. I think the end result is pretty good, certainly much better than the hotch, potch of different equipment it replaced, but I'm sure there's "more" out there somewhere.
|
|
rickcr42
Fully Modded
Rest in peace my good friend.
Posts: 4,514
|
Post by rickcr42 on Jul 6, 2006 22:13:57 GMT
Trying to build a system on "I want the best" can be an expensive lesson in system matching,one I luckily learned at a young age.
You really can't take the "everything best" path to good sound then just slam it together and call it a super system just like you can't just toss "best parts" at a design and call it great just because or build all the so clled "best design blocks" and have a truly good end product and that is why most fail-the rush for "what is the best"
Taking that to DIY projects I notice damn near everyone is locked into the same parts for every amp because they are accepted as being the best but then they take it a step further and use these "best" parts for ALL circuit positions without any thought at all to where and what that part will be actualy doing.
Is the "best" audio signal coupling capacitor also the BEST RF bypass Cap ? Or power supply cap ? Or digital coupler ?
Is the "best" sounding series resistor the best choice for a precision filter network ? For high current power supply duty ? As a DAC IV resistor ? As a plate resistor ?
How about the "best" pot.Will that great sound also mean great precision in a circuit requiring THAT over listeining ?
obviously not so why would anyone think trying to add the "best" circuit PARTS together will make the BEST device ?
The new wave designer is one who scans the various audio forums,pulls parts from threads where the end results is a "best" something then they add it all together in some mongo product and call it a design.
Old school you start out with the premise of "I want to accomplish X" then design each stage to work with the previous one after which you then figure out which powering method is best with that particular device
No "here is the best power supply so I can just plug it in to anything and proclaim success",no "this is the best buffer and it will drive all things better than any other no matter the load" and no "this is the worlds best all situational opamp ever created' so why many attempts at all out "every feature known" DACs designed by hobbysists are lacjking in direct comparison to some very simple far cheaper designs.
Can be a bitch when you beleive you can to build something then find you not only save a boatload of hassle but many times loot as well on what turns out to be a better design because it was actually engineered for the specific duty rather than designed by commitee,popularity of parts or conglomerate of threads on DAc design.Personally I have been building my own DACS and ADCs for a while now but if I ever have the urge to go up to 24 bits I will not go the DIY route but right to the K&K Rakk Dac or maybe the Hagerman Chime though likely the K&K due to the many output options and open ended design.
the DAK you are interested in may well turn out to be a real bargain but for that price neighborhood I would want an audiophile tested tried and true design that has shown some staying power against competition with any reviews by ears with no association or connection to the design threads.
Any involvement at all means any judgement is couded even if by accident because you WANT IT to sound good and is not valid until directly compred to the competition over time-established and proven
|
|