Cleaning up USB power (capacitor?)

[Nazo]

Ok, so I'll admit I'm a DS gamer, and from time to time I plug in my quality headphones since, well, I don't really have anything cheap anymore. There's nothing I can do about the crappy sound quality most software uses (and the fact that the DS itself just doesn't seem to be capable of producing TRULY high quality sound -- though I find that Ouendan is quite acceptable.) One annoyance though is that when I plug in a power source it adds a LOT of noise to the sound. Very poor design on Nintendo's part in this respect... Short of opening up my DS and performing mods on it though, I don't see myself doing much about this. However, I can surely do something about the other end of it however. I used to have a Black & Decker pocket inverter that also had a USB slot for powering USB devices on the go. I found that the battery power from this device was so clean that when I plugged my DS into it it would add no noise that I could hear (though I am a little dissapointed I might add that my motherboard and PSU aren't able to provide power a little closer to clean than this then. I can forgive that cheap little adapter Nintendo provides for charging the DS, but the PC requires higher standards... Apparently USB isn't terribly picky though, so they allow more leeway than is beneficial here.) Luckily I have a little PCB with an input and output jack originally intended to be for adding an external battery to the DS (which didn't work anyway) so adding on something like this without modifying the DS or the cables will be simplicity itself since I need only modify that.

As I understand it, the normal way to clean up an incoming power source is to add a capacitor to it. I know this smooths out voltage changes especially. Perhaps this might be enough though? Or is my biggest problem RFI or something? I tried looping a USB cable through a toroid for this, but it didn't make a difference, so I'm kind of inclined to think it's just an unsteady supply forcing it to take a bit of punishment.

My main question here would be just what sort of capacitor I should use if this is the best solution and do I need one for both sides or just one? I know you can usually use quite a range for this sort of task, but I don't want to break my DS at the moment, so I'd like to make sure I use something safe. Any suggestions here if a capacitor is the correct way to go? The DS is rated for 5.2V, but considering how much you see out there of things using USB to power it I'm guessing it has a pretty fair bit of tolerance since USB specs say devices should allow it to drift down to as low as 4.4V, so assuming it actually supports USB specs as these things should imply, a small drop should be ok.

 

[aL Mac]

You have a lot of unrelated information to your problem in your post so it is kinda hard to sort through... but...

If you wanted to create a filter to filter out noise from your power all you need is a small capacitance. Maybe 0.01 uF that's 0.01 micro Farads. Place it between the power and ground. You would probably need to pick one rated around 10 V.

It wouldn't cause a voltage drop at all. Since only AC current (the noise) should pass through the capacitor and the DC power should not be affected. Typically, decoupling capacitors are supposed to be placed as close as possible to the device that needs it, but in this case it may be hard to do that if it is headphones. When we say as close as possible, it usually refers to on a PCB you would want a decoupling cap as close as possible to the +VCC on the IC that needs clean power.

http://en.wikipedia.org/wiki/Decoupling_capacitor

 

[Nazo]

You have a lot of unrelated information to your problem in your post so it is kinda hard to sort through... but...

There is method to my madness believe it or not. Some of the seemingly irrelevant info is because I wanted to be sure about certain points (such as voltage.)

If you wanted to create a filter to filter out noise from your power all you need is a small capacitance. Maybe 0.01 uF that's 0.01 micro Farads. Place it between the power and ground. You would probably need to pick one rated around 10 V.

Ok, this is the key thing I needed to know. I did a bit of digging around and it seems a lot of things use 0.01uF for this as well, but I wasn't sure as it's hard to find anything very specifically on this subject.

EDIT: Actually, I think I may have seen 0.1uF. Would that be too excessive? I was hoping I could salvage something, but so far haven't had any luck.

Typically, decoupling capacitors are supposed to be placed as close as possible to the device that needs it, but in this case it may be hard to do that if it is headphones.

Actually, I'd say this would be working pretty effectively anyway. I know it is capable of running clean enough not to bother me if I provide a clean power source (eg a battery powered circuit,) so it's not an issue of having an external power source so much as the noise introduced by that source. It strikes me that removing that noise can't exactly be bad for the DS itself either. Oh, I'm sure its circuits are designed to handle a bit of leeway on the input, but there's no harm in making it unnecessary for them to do so as this surely preserves their lifetime some small amount anyway (plus I already know it isn't implementing such a simple basic filtering system or I wouldn't be getting this noise in the first place.)

BTW, one quick question. In my googling around I got the impression this is supposed to be connected in parallel to the power source? Is this correct? I just want to be sure not to toast anything, lol. Also, you say the noise is caused by essentially AC? So I presume I should connect it "backwards" in that the negative should be on the positive input such that only incorrect flow will go through it?

Sorry, my electronics knowledge is rather hit and miss and capactors are my weakest point...

 

[Mohonri]

Simply putting a capacitor between the power and ground lines won't be enough. Since that 5V comes from the power supply, adding a capacitor will only (very slightly) supplement the already-large capacitor in the PSU.

A more effective approach would be to add a diode upstream of the capacitor, like so (forgive my ASCII circuits):

+   ---->|--+---    +
            |
from        =    to DS
PC          |
-   --------+---    -

Just choose a diode that can handle the current your DS expects, and can withstand a reverse voltage of something higher than 5V. As for the capacitor rating, aL Mac is correct--get one rated for 10V (or more). However, I believe aL Mac mixed up decoupling capacitors (which don't apply to this situation) with smoothing capacitors. In this situation, the bigger the capacity (in Farads), the better. If you've got an old busted PSU and a soldering iron, you can even get something way bigger than you need.

 

[Nazo]

Well, I guess I'll be using this with the adapter that came with the DS as well. It too adds a lot of noise. This shouldn't be surprising since the PC which probably has a far cleaner power than anything a tiny little travel adapter can provide is still showing some noise.

If I were to do both, would it harm anything? Especially since the USB isn't actually straight from the PSU -- it goes through the motherboard's circuitry which also seems to have things in effect to control capacitance considering that it won't even allow anything to use more than 500mA (I used to have an external harddrive that required more than 500mA so was no end of troubles for me since it had to be powered from two seperate busses as the system just would not allow it to draw more than the official USB standards allow. Thankfully my new one can be powered via one bus.) Also, if a higher rated capacitor is best, any suggestions, or does it not really matter much just so long as it's a decent rating? I do indeed have access to an old busted PSU and I already have my eye set on one of the diodes in it that looks like it can not only handle a lot of voltage, but it could probably even handle a lot of capacity (likely far more than this application could ever provide since 500mA @ 5V or less should be the maximum even theoretically possible from the USB and 450mA @ around 5.2V the maximum from the adapter that came with the system.)

On the PSU, if I stay a very safe distance from the gigantic 200V capacitors, I see some 10uF capactors that look about ideal for the small casing and some 470uF that might be a little tough to fit in there. I noticed that they have a 47uF on the mains, but it's one of those plastic encased ones that get quite large and I'd rather not use that since it would be sticking out and hanging off the DS in a very annoying manner... One thing I did notice though is the capacitor across the mains has a 470 Kohm resistor across it. Does that work any better?

BTW, thanks for your patience. When it comes to most tasks in electronics I know what I'm doing, but for some reason I've just never really been able to wrap my head around exactly how capacitors are used.

 

[Mohonri]

The resistor across the capacitor is a bleed resistor, so when it's turned off, the capacitors eventually drain off residual charge. You could put one in your circuit, but it wouldn't help the noise issue. It's important to distinguish between the voltage rating of the capacitor and its capacitance. The voltage rating only needs to be about 10V or so. The capacitance is the "more is better" value. Generally, it's a tradeoff: pick two of the three: small size, high voltage rating, high capacitance. A higher voltage rating means you either give up capacitance or get a bigger capacitor. A higher voltage rating doesn't hurt, but it doesn't gain you anything in this application.

(as an aside, capacitors are generally used in two ways: either in series with an AC signal or in parallel with a DC signal. The first is called AC coupling, and it allows two parts of a circuit to communicate or pass signals even if they are on different power supplies or expect a different DC offset. It's also a high-pass filter--high-frequency signals pass through, but low frequencies, including DC, are blocked. The second, which applies in this situation, is DC coupling or voltage smoothing. If the incoming power is noisy, the capacitor soaks up some of the charge when it's too high and releases charge when the power is too low, thus smoothing out the incoming power. It's also known as a low-pass filter, since the low frequencies (including DC) pass through, but the high frequencies get soaked up.)

 

[Nazo]

Ok. I found a small 100uF 16V capacitor in there. If more is better, it should be more than sufficient for this considering that 47uF was enough for a mains on a PSU. Oh, and I understood about the voltage ratings. The most convenient capacitors I have on hand are all 10V or higher, so I wasn't terribly concerned about that.

BTW, I must admit to trying to soak up a bit of this, lol. I'm building a couple of CMoys soon to go between my new headphones and my MP3 player, DS, and etc and the use of capacitors in it is the main thing I didn't really understand. Lol, well, I still don't fully, but I'm starting to get some of the idea here, especially with its properties in DC versus AC.

Anyway, I just have one last question to be sure on. Which way should I turn it? With the negative terminal on the positive side, or do I still have it on the negative side?

 

[mtrupi]

Anyway, I just have one last question to be sure on. Which way should I turn it? With the negative terminal on the positive side, or do I still have it on the negative side?

The negative voltage goes to the negative terminal and positive to positive. Do not put the reverse polarity on a polarized cap as they can literally explode. A USB port may not have enough oomph to do anything quite so dramatic. Also, in general you need to be concerned with surge current ratings. Again, the USB application may not be a problem but others are. Finally, just the addition of a capacitor may not solve your noise problem but it's a relatively easy place to start.

 

[Nazo]

Ok, thanks for all of your help. I guess I'll have to buy a diode since I can't seem to find one that isn't mounted so well that I have to damage it to get it out (I don't have any desoldering braids, so have had to do my desoldering the hard way, which doesn't work too well) so I guess I'll have to try this later. Well, I need a couple of volume controls apparently since I changed my mind on the CMoy design AFTER ordering the parts (always the best time.)

Plus, apparently the portable adapter isn't giving me noise after all. At least, not at the moment. Can't really explain that since I swear it did before and I don't think anything has changed (it's even still in the same plug. Actually, if anything it should be worse since I think it's on the same circuit as a window A/C...)

Anyway, thanks for your help and patience.

 

[Mohonri]

The negative voltage goes to the negative terminal and positive to positive. Do not put the reverse polarity on a polarized cap as they can literally explode.

That's true, you know. Back in college, I was a TA for a lab (lab assistant?), and one day, one of the students asked what would happen if we hooked up a capacitor backwards. Having never done it myself, I myself was curious, so I told him "Go ahead and try it." In retrospect, it probably wasn't the smartest thing I've ever done. They hooked it up to the bench power supply and gradually turned up the voltage until it popped. Now, this was a small capacitor--perhaps 6 or 7mm in diameter. But the pop it made when the electrolyte boiled was rather satisfying and surprisingly loud.

 

[aL Mac]

0.1 uF would work also. Either is typical

I wouldn't use a diode. I disagree with the comment about adding a diode and that it would only supplement an already large capacitor. The reason I disagree, is that the 0.01 uF capacitor would be to filter out higher frequency noise, not noise from the AC power. The diode would cause a voltage drop of about 0.7 volts which is undesireable. If I were you, I would try using a 0.01 uF or 0.1 uF capacitor between power and GND (I suppose that's parallel as you stated), if I added anything else it would be a larger capacitor such as a 3300 uF to filter out any additional ripple from the AC adapter.

 

[Nazo]

Hmm. Well, first of all, I've discovered to my annoyance that even though I specifically stated that the AC adapter was providing a lot of noise, apparently it is not currently. I hope this isn't temporary, but I could have sworn it was before, so maybe it will go back to it (I swear I haven't changed anything, lol.) So I guess that's a none-issue at this point.

Anyway, I don't currently have any really low rated capacitors. The lowest I've found on hand so far is 10uF. I tried a 100uF since he said bigger is better, but it's not really helping on the USB. Perhaps I do need lower for it to work. I can't really say right now since I can't really properly test that. I won't be paying a visit to my local RadioShack for a while and since I've discovered the main AC adapter seems to not be putting out noise atm I may just not even bother with the USB. Actually, on that subject, my multimeter tells me that the AC adapter is putting out 5.25 volts, which is just the tiniest amount over the official rating. By comparison though, I think the last time I tested my USB it was a smidge low (though still well within the 5% standard.) Of course, 5% for USB means it can go as low as 4.75V versus 5% for the DS being about 4.94V. The DS is a rugged little device made to take a lot though, so I wonder if the noise I'm getting might be some voltage converting kicking in to compensate? I figured the DS was still way over what it needed since it runs off of a 3.7V lithium-ion battery, but I admit to a bit of curiosity. Of course, it could just be that the USB is that noisy, I just find it odd that it comes out so much noiser than an AC adapter designed for travel convenience rather than quality...

 

[aL Mac]

back up.. let me restate all of this.

First off about the diode:

• It will not have any desirable effect, the output voltage will always positively bias the diode.
• It will have a negative effect by causing a voltage drop of 0.7 volts
• Do not use a diode

About the small filtering capacitor:

• Bigger is not better, you need a small value to filter out high frequencies. If you use a big capacitor it will not work
• Typically you want a 0.01 uF capacitor or 0.1 uF capacitor between power and ground. Just a cheap ceramic cap will work, you do not need an electrolytic capacitor. A ceramic cap has no polarity.
• A big capacitor would help a seperate phenomenon called ripple which occurs from rectifying the AC signal to a DC signal. If the capacitor was not big enough in the power brick itself, a big capacitor could possibly help, maybe 3300 uF in parallel with the other smaller capacitance.
• The capacitors only filter alternating current. Therefore, the DC component, or offset, which is your power, will have no voltage drop associated with the capacitor. The terminology of capacitors "letting high frequencies communicate" is really just confusing.

Now, about your digital multimeter:

• It is not made to be able to measure high frequency AC signals
• It mesaures RMS voltage, which is 'root mean square". You can think of it as an average voltage.
• Therefore, a 5 V signal with an AC signal on top of it, will not truely show the AC signal on top of the 5V.
• To measure the noise on the DC power supply you would need an oscilloscope, which is beyond you technical know how and budget.

Lastly, it is difficult to know if you have a problem with noise from your power supply or not. It seems like you are still trying to figure out exactly what is the problem. There are many things that could affect your results and often problems with noise are never straight forward. Good luck, and I hope that whatever comes of your project you at least learn something =]

-Al
(Computer Engineer)

 

[Nazo]

Now, about your digital multimeter

Oh no, I understand all of that stuff. In fact, it didn't even occur to me to try to look for such a signal using the multimeter of all things... I was merely saying that while the AC adapter easily provides within the industry accepted 5% range from the DS's specs, USB provides well outside of that range. Were it to have to do voltage conversion or some such thing, perhaps that could add noise? Well, it probably doesn't really have to since it almost surely is going down from that considering how the system is capable of being powered from 3.7V after all, but it was just a thought.

To measure the noise on the DC power supply you would need an oscilloscope, which is beyond you technical know how and budget.

With all due respect, that's getting just a bit on the insulting side now. Though my electronics experience has its holes, I'm not a moron... I will admit to the budget limitations, I just want to point out that despite the holes, I still do have a decent amount of experience.

Lastly, it is difficult to know if you have a problem with noise from your power supply or not. It seems like you are still trying to figure out exactly what is the problem. There are many things that could affect your results and often problems with noise are never straight forward.

Agreed. Actually, I just tested a theory. Up until now I've been using my headphones which just can't get enough out of the DS (in particular I've always found my DS to output a too low volume level, but some of that may be it being designed for cheap 8-ohm sets? I don't know though, my Shures are 16-ohm, but I'll admit they seem to require more power for some reason despite this...) I'm having troubles with those Shures though as I broke one earpad (or should that be canalpad? lol) so currently am being too cheap to buy a set of five pairs (which means ten total!) to replace one and everything I've come up with to get a good seal so far is massively uncomfortable and I've been primarily using my headphones for a good while now (ever tried SR60s on a DS? Lol.)

For testing purposes though, I plugged the E2Cs in for a bit and ignored the pain in my ear canals long enough to test and guess what? The noise only shows up when I have the DS plugged into speakers/soundcards/etc (I wonder how it will do when I get CMoys? Their shockingly simplistic design may really shine here.) This makes me think my problem may be more fundamental and harder to fix -- namely, RFI rather than simple voltage issues. Maybe a grounding issue? Oh well, my new Denons don't require as much power to drive (despite a similar and even overall somewhat lower response curve and a very similar impedence curve) and while they are a bit quiet without an external amp due to the DS's low output even at maximum volume, it's still a decently acceptable volume and I can put up with that for now. Hopefully the CMoys solve that issue anyway. I'm building two so one can stay at home and the other can be more portable anyway, so I will likely always have one on hand. I'm willing to bet they don't pick up or magnify noise much beyond what is in the actual sound itself.

 

[Mohonri]

Heh... You want duelling degrees? EE here! lol

My reference to a series capacitor passing high frequencies was intended only for perspective, since Nazo expressed some confusion. Certainly that application doesn't have anything to do with this situation.

al Mac is correct in stating that a small ceramic capacitor is better suited to the application, due to the lower ESR. (Don't worry about the technical details--it just means it soaks up high-frequency noise better when placed in parallel) However, I still stand by my recommendation to add the diode. Diodes (or rectifiers) with a drop of under 0.5V are easily found, and Schottky diodes can have even lower voltage drop.

All this is descending into a discussion about unimportant details, however. I agree with aL Mac--why are you getting noise in the first place? In all likelihood, the power, once it's inside the DS, is almost certainly regulated down to 3.3V or something similar for use by all the electronics, including the audio hardware. That voltage regulation should take most, if not nearly all, electrical noise on the power supply. I can't help but wonder if there's another problem rearing its head. Since the problem seems to be intermittent, I wonder if it's something more mundane, like a loose connection. I know you're loath to open up your DS, though, so it's not very fair to ask you to crack it open and look at all the connections.

 

[Nazo]

Lol, I think I caught you with my post while you were still writing yours. Yes, when I investigated further it seems this may not be noise from the power exactly directly. Oh it's ultimately coming from there as I only get this noise with an external power source plugged in, but then too the noise is also only present with an external thing like my speakers or soundcard. And I'm using good speakers too, so they don't add noticable noise until I turn the volume control up to a level that, if anything was playing, would either blow out my headphones or my ears (or more likely, both...) Actually, once I thought about it a little more just now, I realized this is probably a ground loop issue. I've had the misfortune to see those before and not coincidentally also with the computer (though different hardware at the time.) At the time it was video, but in the end it's an all too familiar issue.

BTW, since we are comparing degrees/experience *raises hand*. Ah... Lowers hand. All self-taught here, lol. I can't even stand to read books about it. But yeah, things like this are always a learning experience. I understand capacitors a LOT better now than I did before.

EDIT: I just remembered I had one of those cheap "Boostaroo" headphone splitter/amplifiers lying around and dug it out. It uses a design similarly simplistic to the CMoy and I get no noise with it. This sounds to me more and more like a ground loop issue.

 

[aL Mac]

I wasn't trying to insult you about the oscilloscope, I just assumed you didn't have one. I also figured that they were too expensive to be worth it for this one application.

I have a computer engineering degree. However, I took several circuits, electronics, signals, digital electronics courses etc. Computer Engineers at my school took most all the core EE courses, we just get less electives and specialize mainly in digital electronics, sequential logic, and computer design

Anyway.... can you explain to me why he would need a diode? I can't say I'm an expert or anything, but thinking of a full bridge rectifier it would already have a diodes in place to rectify the signal, with the smoothing capacitor, the output voltage would have some negligible (hopefully) ripple, but it's average output should be near (I forget what it is let's just assume) 5 V. So even if you had terrible ripple and it went from +4.5 to 5.5, what would the diode do? The current is still flowing in the same direction always, because the voltage polarity will not change after the rectifier

 

[mtrupi]

OP, there's is a lot of confusing information and misinformation here. Are you trying to to power the DS from a USB port? The USB spec allows for a nominal 5V at 500mA, is that enough for the DS? The power from the USB port will have gone thru a power switch device and it will already have a significant capacitance and probably ferrite beads as well. There is a max capacitance (10uF?) the spec allows a device to have without surge current limiting capability. Putting a larger capacitance may or may not cause voltage droop and excessive current spikes. Even if it does it may not cause any problem for you. It's hard to know the cause of your problem but since you hear the noise, it is a relatively low frequency. Power circuits are low impedance by nature so any reasonable capacitance you put there probably won't help. The diode idea likely won't help either. There's just not enough information to say I know what the problem is and how to fix it. If you want to try things out by all means do it, just do it safely. If nothing else it will be a learning experience.

 

[Nazo]

Officially USB is supposed to be ok. If you pay a visit to any local department store and look at their DS stuff you will find a minimum of one kit with a USB power adapter (in particular they like to use this for car power. It's easy to find a USB adapter that can be rebranded as these are massively common, so they just grab one of these and throw in a USB cable for the DS.) Well, really I agree with Mohonri in that the DS most likely doesn't need such a high voltage really. Remember, this thing can be powered by a single cell lithium-ion battery, which means 3.7V. To use higher it would actually have to use voltage conversion which is a highly inefficient method of powering a device from the perspective of the extreme efficiency they were focusing on (this thing uses shockingly little power considering just how many hours it can get from a 1000mA battery...) I might add here that when I plug it up via USB it is able to not just run off of the USB power, but it can even charge the battery while running.

There's just not enough information to say I know what the problem is and how to fix it.

Maybe, but, read my last post. I think you'll find there is enough information now to narrow things down a lot.

 

[mtrupi]

Maybe, but, read my last post. I think you'll find there is enough information now to narrow things down a lot.

You are drawing the conclusion it is a ground loop problem? I can't make enough out of what you wrote to do the same.

 

[diredesire]

Oh no, I understand all of that stuff. In fact, it didn't even occur to me to try to look for such a signal using the multimeter of all things... I was merely saying that while the AC adapter easily provides within the industry accepted 5% range from the DS's specs, USB provides well outside of that range. Were it to have to do voltage conversion or some such thing, perhaps that could add noise? Well, it probably doesn't really have to since it almost surely is going down from that considering how the system is capable of being powered from 3.7V after all, but it was just a thought.

I think you can rule out power. The problem is your audio output stage, IMO. The voltage conversion should cause no issue, as it's probably put through a vreg, which is going to provide just as clean of a supply provided the input voltage is within spec (which it should be).

Agreed. Actually, I just tested a theory. Up until now I've been using my headphones which just can't get enough out of the DS (in particular I've always found my DS to output a too low volume level, but some of that may be it being designed for cheap 8-ohm sets? I don't know though, my Shures are 16-ohm, but I'll admit they seem to require more power for some reason despite this...) I'm having troubles with those Shures though as I broke one earpad (or should that be canalpad? lol) so currently am being too cheap to buy a set of five pairs (which means ten total!) to replace one and everything I've come up with to get a good seal so far is massively uncomfortable and I've been primarily using my headphones for a good while now (ever tried SR60s on a DS? Lol.)

Your theory really leads me to believe it is your output stage. The impedance matching that you suggest has some merit, except I'd actually guess it was designed for a 16-32 ohm load, rather than an 8 ohm load (driving an 4-8 ohm load is difficult). Your Shures might be a little current hungry, which is why they may sound "lacking" in terms of "power." (all loose terminology, here). Also, if you want a spare earpad, shoot me a PM, i probably have some that will fit.

For testing purposes though, I plugged the E2Cs in for a bit and ignored the pain in my ear canals long enough to test and guess what? The noise only shows up when I have the DS plugged into speakers/soundcards/etc (I wonder how it will do when I get CMoys? Their shockingly simplistic design may really shine here.) This makes me think my problem may be more fundamental and harder to fix -- namely, RFI rather than simple voltage issues. Maybe a grounding issue? Oh well, my new Denons don't require as much power to drive (despite a similar and even overall somewhat lower response curve and a very similar impedence curve) and while they are a bit quiet without an external amp due to the DS's low output even at maximum volume, it's still a decently acceptable volume and I can put up with that for now. Hopefully the CMoys solve that issue anyway. I'm building two so one can stay at home and the other can be more portable anyway, so I will likely always have one on hand. I'm willing to bet they don't pick up or magnify noise much beyond what is in the actual sound itself.

The issue here is that the DS is not designed to power speakers. You mention that you are powering a pair of Denon speakers un-amped with your DS? This is a little bit ... crazy. As far as the CMoy improving it, this might be the case, but the CMoy isn't really designed to power speakers, either. You'll get a bit of gain, but not really any circuitry that would be suited to drive speakers. The low impedance of speakers (4-8 ohms) is going to be an ADDED stress on your output stage(s). The CMoy would effectively (IMO) act as a buffer to your DS output, which might save you there. The CMoys won't "pick up" noise, but they sure will 'magnify' it (it's gain!). It'll sound the same (quality-wise), just louder. You may have some noise filtering, but it's not going to be what you're looking for.

Lol, I think I caught you with my post while you were still writing yours. Yes, when I investigated further it seems this may not be noise from the power exactly directly. Oh it's ultimately coming from there as I only get this noise with an external power source plugged in, but then too the noise is also only present with an external thing like my speakers or soundcard. And I'm using good speakers too, so they don't add noticable noise until I turn the volume control up to a level that, if anything was playing, would either blow out my headphones or my ears (or more likely, both...) Actually, once I thought about it a little more just now, I realized this is probably a ground loop issue. I've had the misfortune to see those before and not coincidentally also with the computer (though different hardware at the time.) At the time it was video, but in the end it's an all too familiar issue.

"Good speakers" is a very gray term, IMO. Even if they are much more efficient, they will still present a load to the DS that is beyond it's capabilities. The fact that you added a boostaroo suggests that the stage is fine (considering that the boostaroo is effectively buffering (and taking the large load off of) your DS stage. I don't think it is a "ground loop" issue, just an impedance/design mismatch.

The best thing you can do right now is plug the DS in to an amp that is dedicated to driving the speakers, or just use a boostaroo as a buffer.

Also, are these speakers strictly stand alone, or are they powered? If they are powered, they probably have a rudimentary amp (probably not so dis-similar to a CMoy, but with a higher current (maybe?) op-amp.

 

[Nazo]

The issue here is that the DS is not designed to power speakers. You mention that you are powering a pair of Denon speakers un-amped with your DS? This is a little bit ... crazy. As far as the CMoy improving it, this might be the case, but the CMoy isn't really designed to power speakers, either. You'll get a bit of gain, but not really any circuitry that would be suited to drive speakers. The low impedance of speakers (4-8 ohms) is going to be an ADDED stress on your output stage(s). The CMoy would effectively (IMO) act as a buffer to your DS output, which might save you there. The CMoys won't "pick up" noise, but they sure will 'magnify' it (it's gain!). It'll sound the same (quality-wise), just louder. You may have some noise filtering, but it's not going to be what you're looking for.

I'm not powering speakers via the DS, the Boostaroo, or even the soundcard. They are powered speakers already and don't need help. The Denons are headphones as are the Grados and the Shures are IEMs, all of which are anything but speakers. Also, the point I was making about the CMoys and Boostaroo versus the soundcard and the speakers is that both the soundcard and speakers are FAR more susceptable to RFI. In fact, the Boostaroo and CMoy should only be able to amplify any noise AFTER it gets into the audio signal itself rather than before. Since I get no noise in the audio signal without an amp, I know these should be ok. Not a big surprise really. They don't have a bunch of complex parts that are suceptable to RFI like the speakers and soundcard both do. Those two clean up input power, but I'm willing to bet they can't really be designed to really clean up a bad input from the audio line itself.

Also, are these speakers strictly stand alone, or are they powered? If they are powered, they probably have a rudimentary amp (probably not so dis-similar to a CMoy, but with a higher current (maybe?) op-amp.

Far more complex than that. I have a full desktop set, the Klipsch Promedia 2.1. They obviously clean up the mains and that sort of thing (the SNR is amazing. I have to turn the volume up to about 3/4 to hear noise and even then I can only hear it with my headphones normally -- I might add here that I've never had even quiet music with the volume control set higher than 1/4, so 3/4th is rather high...) However, can they really do much with a bad line-in? Are they even designed to? I doubt it. They aren't cheaply designed though -- not by a long shot.

I don't really listen with the speakers directly much though these days. Most of my listening is with headphones or IEMs. So once I have the CMoys built, it will be like listening through the Boostaroo, but with a good gain and a quality opamp (well, I'm going with the Burr-Brown/TI OPA2132PA and will be trying out the OPA2227PA as well, so admitedly some may disagree about the sound quality, but it sounds like it will be quite suitable to my needs and a lot of people do enjoy them.) To that end, I'm not really terribly concerned over this anymore unless for some weird reason the CMoy actually SHOULD pick up noise where the Boostaroo does not, but I find that rather highly unlikely...

 

[diredesire]

I'm not powering speakers via the DS, the Boostaroo, or even the soundcard. They are powered speakers already and don't need help. The Denons are headphones as are the Grados and the Shures are IEMs, all of which are anything but speakers. Also, the point I was making about the CMoys and Boostaroo versus the soundcard and the speakers is that both the soundcard and speakers are FAR more susceptable to RFI. In fact, the Boostaroo and CMoy should only be able to amplify any noise AFTER it gets into the audio signal itself rather than before. Since I get no noise in the audio signal without an amp, I know these should be ok. Not a big surprise really. They don't have a bunch of complex parts that are suceptable to RFI like the speakers and soundcard both do. Those two clean up input power, but I'm willing to bet they can't really be designed to really clean up a bad input from the audio line itself.

I realize the DS/Boostaro is not a power source, I should have been more clear. I meant using the boostaroo/DS as an input to your speakers. (If you understood this, your second sentence made no sense to me). As far as RFI, if it was the speakers, going with your "complex parts" theory, you should be able to eliminate any RFI issues that aren't taken care of by the circuit design with some ferrite cores. (I know this was mentioned above). The soundcard is a legitimate suspect, though, as it is providing a signal in a "noisy" environment. (BTW, I think it is obvious that the CMoy and Boostaroo can only amplify signals that are in the input.. ).

I should mention that I strongly believe susceptibility to RFI is more or less the same, it is just the degree of which each component is exposed to RFI. The soundcard is in a noisy environment, and the speakers may be less shielded. A poorly implemented CMoy can easily pick up garbage (fluorescent lights, fans, etc) just like a sound card. I think you are overplaying the simplicity of the CMoy/Boostaroo in this regard.

As far as bad input, it can be cleaned, depending on the noise. The amplification stage in either component (depending on design) can filter out noise common to both inputs (assuming differential). (CMRR)

Far more complex than that. I have a full desktop set, the Klipsch Promedia 2.1. They obviously clean up the mains and that sort of thing (the SNR is amazing. I have to turn the volume up to about 3/4 to hear noise and even then I can only hear it with my headphones normally -- I might add here that I've never had even quiet music with the volume control set higher than 1/4, so 3/4th is rather high...) However, can they really do much with a bad line-in? Are they even designed to? I doubt it. They aren't cheaply designed though -- not by a long shot.

Far more complex than what? You're going to have a power amplifier for your speakers, and then a headphone amp/output stage that is probably about on par with your CMoy? If we assume the line-in is bad, are we placing the blame on the DS, then? Like the other posters above I'm really not sure what you're trying to analyze, here. I'm not being condescending when I say: Can you post a summary of what we know so far? I am seriously reading loops above. What configuration specifically gives you noise?

From above you say it happens only when an external source for the DS is plugged in, and also when you are powering speakers/soundcard? (by power I mean giving an input). Does this mean you are feeding your DS's headphone output into your line in on your sound card which is getting fed back to the Promedias (I can't see why else you would be hearing noise... are we even talking about DS specific sound?)

Sorry I'm so confused, but you have a lot of information in your posts that is unrelated, and I am curious to help you find what's up.

I don't really listen with the speakers directly much though these days. Most of my listening is with headphones or IEMs. So once I have the CMoys built, it will be like listening through the Boostaroo, but with a good gain and a quality opamp (well, I'm going with the Burr-Brown/TI OPA2132PA and will be trying out the OPA2227PA as well, so admitedly some may disagree about the sound quality, but it sounds like it will be quite suitable to my needs and a lot of people do enjoy them.) To that end, I'm not really terribly concerned over this anymore unless for some weird reason the CMoy actually SHOULD pick up noise where the Boostaroo does not, but I find that rather highly unlikely...

I suggest the 2227, I feel it's a little tighter than the 2132/4. If you are going to run your IEMs, I suggest dropping your gain to about 2-3. You don't necessarily want "good gain," but you do want a quality amplification stage. If you decide you like building CMoys, I'd be interested to see if you come up with any interesting circuit layouts
I've been working on a few recently, but just for fun. The CMoy is a good intro amp, but if you like it I really suggest upgrading in a hurry
(Especially if you want to open up your Grados).
http://students.washington.edu/jdeng/minimoy3.jpg

 

[Nazo]

What I was saying is that, as you put it, the soundcard and speakers were providing a far messier environment than the CMoy and Boostaroo provide. The noise only appears when using them. The CMoy and Boostaroo have a far simpler environment and aren't picking up this noise. I'm pretty sure this is a ground loop issue still as everything has been quite consistent with what that does with the computer's setup being rather messier than that of, say, a DS or a Boostaroo or CMoy. (Then too, admitedly it could just be related to something about the motherboard itself -- maybe solid state capacitors are harder to set up as well or something? I wouldn't know, I just know that if I ever decide to overclock I have a lot more confidence that this board will hold up regarldess of any noise that may appear when using USB to charge.) Anyway, when I listen using the Boostaroo or CMoy there is no noise and now that I have these I really don't have to run through the soundcard or speakers. Problem solved.

As for the CMoy, so far I'm not really suceeding in making my own. In the meantime I picked up one that someone was selling on eBay. Actually, I got a pretty good deal. It comes in the better looking Penguin case (as opposed to the Altoids or whatever with the bright colors that stand out when you have the thing out in public) and it uses the OPA2227 opamp. I swear though, this thing looks like it was made by a professional company rather than a hobbyist. All of the soldered wires have been shrinkwrapped to help protect those solder points. It's actually probably based on that MiniMoy design like you just showed, only the audio connectors are wired so they can go to the case seperately and it has the volume pot (this makes the base curcuitry itself even smaller overall though.) I'm quite impressed really... (This isn't that Walking Wolf or whatever their name was that you normally see on eBay -- I was going to get one of those and just switch out the opamp, but couldn't find one with a decent buy it now price and then I lucked out and someone put up this really nice one for a REALLY good price considering how well made it is.) I was hoping when buying that I'd be able to look over their amp and basically copy from it so that whatever I did wrong I wouldn't do again, but there's no way I can copy this thing... Lol, I'll try again, but I think I'm not going to be doing anything on par with this sort and I'm willing to bet I won't come up with new methods or anything that would be of any use to you.

Oh, and I'm less concerned about my IEMs than my headphones. Forget the Grados. I picked up the Denon AH-D1001S a bit ago and have burned them in and while at first I had intended to use the Denons only in a noisy environment I now am considering trying to convince someone to buy the Grados at not too much of a loss (considering that I already invested in the L-Cush earpads even.) I LOVE these Denons and will be using them the most often at the times where I would need an amp. I have not found one thing that doesn't sound better on the Denons now that I've burned them in (only at first did the Grados sound better.) I'm a little annoyed actually because thanks to Amazon I got these Denons for less than the SR-60s + L-Cush added up to (which is actually pretty nice since normally people would be more inclined to get the Grados for the simple reason that the AH-D1001 and even 1000 is normally closer to twice the cost of the SR-60s...) Anyway, the CMoy seems to be quite sufficient so far. Going further than the CMoy seems a bit much as there's the issue of price for one (I spent enough just getting the headphones already and it seems like most amps cost just as much or even more.)

 

[BallerX]

http://www.minidisc.org/headbanger.html
That will power speakers just fine. Mine does anyway.
If I where the OP I would just get a 5v regulated wall supply from radioshack and just mod a USB cable to plug the DS in. I cant run my amp off my PC line without horrible noise either. So I just use a battery. I run a 9.6v 100mah battery. It last for at last a week of continuous use. It charges in 4 hours. YMMV.

 

[Nazo]

Er, believe it or not, posts after the op do have meaning. The problem has been basically solved. Oh, and my speakers already have an amp and I already have a wall plug for my DS (one comes with the DS) so the USB was just for convenience.

 

[diredesire]

If you're looking for another portable CMoy, I'd love to trade for the SR60s (+ cash, of course)

 

[Nazo]

That I will think about. I mean, the SR60s are good (especially with the L-Cush pads -- though I also tried quarter-modding the normal pads just to see if that helped any) but since these Denons burned in I'm just not really getting any enjoyment from the SR60s anymore. (Man was I shocked too. The Denons are closed sets! They shouldn't sound NEARLY so good even at the normal pricerange, but at the same amount or maybe even a bit less as what I spent on the SR60s after adding the L-cush pads, they are ridiculous.) I was hoping to find someone more local (such as my mother -- she listens to music close to as much as I do, and that's a LOT) but so far haven't really been able to convince anyone to buy headphones that cost more than a pair of Skullcandies. The only thing is that I'm not sure if I do need another CMoy. I now have one good one and I still mean to try to put another together later with hopefully less dissapointing results (of course, at this rate I may not be able to since I managed to mess up that first one somehow.) I'll definitely give it some thought.

 

[diredesire]

No worries, just fire me a PM if you decide. I'd be willing to trade an Alien DAC as well (head-fi for more details).

I love the Grado SR-60s. I have the Alessandro MS-1s, but I'd love to have a "beater" pair to mod as I please.