Hi in a previous video which I'll link in at the end and down below. If you haven't seen it, these Kok Roca Door Rock at Six powered studio monitor speakers. And it's not just these, but most of the popular studio monitors on the market. Hashtag: Not all studio monitors actually produce this low level hissing noise generally from the tweeter.

There is a bit of low-level hum from the speaker, but basically from the tweeter. If you put your ear up close to it with no input signal whatsoever, then it generates this in turn or yes, and this is common across most of your popular brand studio monitor speakers. Some of them can be many thousands of dollars and a common comment on the previous video was that hey, well, what do you expect on these cheap Chinese speakers, right? Well one, they're not that particularly cheap. and as I said, some speakers in the studio Mondas top brand names into the thousands at all still also have this issue.

but hashtag, not all of them and the some people were saying well, this brand doesn't do this top-end brand, you know, really expensive top-end brand. They don't do it and then somebody else goes oh yeah, I've got those Yeah, there's still some noise on there, you know. So like it's just common as mud across most of your popular I Won't say they say the real high-end ones cuz apparently they're you know, much more expensive for a reason. but all of your most popular and largest selling brands UK are case your Yamaha is your JBL's etc etc.

They all tend to have some amount of noise or hiss you know, particularly in the high frequency drive of the tweeter. and it's not really an issue in terms of a signal noise ratio. I mean some might be a lot worse than others and you might be able to hear them if they're in a real silent studio or something like that with no input signal going in, but you know it generally is not a problem. Once you put a signal on there, the signal-to-noise ratio is just so ridiculously high.

It doesn't matter, so it doesn't affect the performance of these things, but it's still there and it's noticeable and it's annoying. So I thought we'd have another quick look at this now. this speak is actually the other one that I've got. The previous one wouldn't notice that we had about I think it was about 250 micro volts RMS noise as measured on the 121 GW here across the 4 ohm tweeter on the K Okay, this other one that I've got actually significantly higher than that.

Let's quarrel about you know, four hundred and thirty odd micro volts and you can. It is noticeably a bit louder. so let me see if I can pick up that noise for you and this road M5 instrument mic that I'm using here. I've got the gain turned up to absolute max on my Sony In Xat camera audio amplifier here and you can barely hear it and that might sound loud, but really I can't hear it from here I've got to put my ear a couple of inches away before I can hear that.

So we tried various techniques in the previous video. I go over them again to try and isolate the noise, but we basically isolated it down to the output power amplifier chip on here. It's basically inherent in the noise floor of that chip. It doesn't matter how to input termination or anything else, but the one thing we didn't do is actually try and isolate the power supply on this.

Now, if you have a look at the schematic, there's three separate supplies for this thing. The first one is plus Minus twenty seven volts, which is goes directly to the low-frequency power driver and that's not the one we're interested in today and that then is I goes into some regulation to transistor regulators there to produce plus minus 15 volts for the Op amps on the front end of this thing which we've determined they contribute very little to the noise of this thing and we can get it to change a little bit like 5% or 10% or something that it was, but the bulk of its come in from with inside the high frequency amplifier which is the TDA 2052 and that's got its own separate power supply over plus minus nineteen point five volts and that's coming directly from a just a standard full wave bridge rectifier on the input directly to some filter Pasteur's and that's it. So I thought hey, you know what if even though it's just you know, it's just 50 Hertz But some people said hey, there could be some bridge rectifier noise or something like that being generated on the high current pulses or something like that. but it doesn't matter, this is noise is down at the quiescent current of the amplifier.

But anyway I thought to answer this question, we actually power the high frequency amplifier directly from batteries and see if the noise goes away, see if it is actually related in any way to the power supply. I Don't believe it is. I Believe it's the internal architecture of the like, the power amplifier stages within the amplifier chip, the topology used, and all that sort of stuff. There's various.

You know you could do a PhD thesis on noise in this one's actually a class A/b amplifiers, but hey, we've got a baseline here. Let's plug in some batteries, see if it makes a difference. And of course you can't just use a lab power supply for this because they're also inherently noisy. so you know really the best way to do it is just power for some batteries so I'll cover together.

Aha A plus minus 18 volt battery rail. the quiescent current of the amplifier the TVA 2052 is only like 40 milliamps so we don't need a lot. We could have even used like 9-volt batteries, but I'll use some triple A's okay. I've got a plus minus 18 volt supply near enough to plus minus our 19.

not sure what the actual because it's unregulated from the transformer. I'm not sure the actual voltage. It's going to be good enough for the purposes. Today's experiment: I'm drawing 21 and a half million from this and I've got a so it's a plus minus 18 volt rail and it's split like that.

Now we're getting nothing on the output. I Think that's because it's in mute mode. whoops, right? So what I'm going to do is get in there and actually short out the cathode of the diode on the mute circuit. I'm going to short that out to ground here and that should enable or unmute the high frequency amplifier.

so you probably expect the current to go up and we'll watch the voltage across the tweeter. So let's give this a go way. Hello, 60 milliamps And that's interesting. A hundred micro volts, so it is certainly lower, but it's probably still going to be there.

Okay, let's try and listen to this now. I've permanently sold it in the the mute mode, so that's not as loud on the Vu meter on the camcorder, and although it's measuring a lot lower at 100 microvolts to my ear, it's still the same type of white noise. It's a little bit lower, but not by a huge amount. Still there, you can still hear it, but it has it has reduced so you know that is reflecting on the Rms noise measurement and I just realized I had the input board which is not being powered but I still have it connected and well, big.

I disconnected it so it's now flapping around in the breeze. so I need to disconnect it from the other end or short the input out. Okay, so what I've done now is I've just shorted across and probably can't see it. but I short across the 10k input resistor down there I still got my bypass on for the mute and I've disconnected this so it's not so nothing's floating around in the breeze.

It's connected directly, short, loop directly through the correct ground, back through to there and we still get a hundred micro volts and well, let's have a listen. Yep, still there and we'll just record that. There we go. So there you go.

That's an interesting experiment. We were able to actually get the noise down using a DC supply here. Granted, it's not the same voltage, but it's near enough. It's like seventeen and a half volts under load.

I've checked that, you know, as opposed to Nineteen Point Five. That doesn't explain the drastic drop in noise that we actually got there. So even if we short the input, a lot of people said, oh, it's grounding and all that ground loops and all that sort of stuff. No, it's got nothing to do with it.

There is noise inherent within the amplifier chip in there. It's just inherent in the topology. But the interesting thing is, we were able to lower it. but I can still hear it.

So really, you know I might maybe do a part three investigating how to clean up the supply. So maybe there is some Junction noise coming from the bridge rectifier diodes or come in, you know, from some you know, coming in from the input line or whatever going through. So maybe you could add some filtering heavy-duty filter in because there are heavy power rails, you can't just put little piss-weak filter in there. It's got to be high current stuff.

Maybe you could. you know you regulate it better be talking about beefy power regulation now in that respect, but that can introduce its own noise and that sort of stuff. But potentially we could put some, you know, passive component noise reduction in there or something like that. But unless they get it down to zero, like, like, inaudible.

Basically I can still hear it I'm not really going to be happy. So anyway, there you go. That was an interesting experiment. It's still there, but it is somewhat reduced.

but still. if I put my ear within 2-3 inches of this thing, I can still hear it. So yeah, what do you do anyway? I Thought that was an interesting experiment. and if you actually look at the data sheet for this chip and the input referred noise and multiply it by the gain of 27 that this thing has, you know it's starting to get close to this.

So it's like the input referred noise. There may be some additional noise internally from the topology or whatever, you know, the class A/b topology in there, and all that sort of jazz. But ultimately, yeah, we've you know, reach the limit of that amplifier chip and most amplified chips on the market like off-the-shelf ones. So if you wanted your own lower noise one, you're really gonna have to roll your own.

and most manufacturers just don't bother to do that. Maybe some of the real high-end ones do. And as I've mentioned before, just be careful about you know, if you're testing this on your monitor speakers make sure it doesn't have an auto mute function. This one doesn't do it.

but a lot of them will actually mute. If you've got no input signal, they'll actually mute the output driver and this one is able to mute the output driver down to inaudible levels you can't hear. It is basically down to zero as you saw on the measurement, but this one, this particular speaker only does that on power up, but there are other models that will actively do that, so it doesn't mean that the noise isn't there, it is there, just sort of masking it. And yeah, that's the whole point, the whole intention why the manufacturers actually add that feature is because they want to use a you know, like an off-the-shelf audio amplifier chip like this which has some noise.

and with the you know, excellent, generally excellent high sensitivity speakers Inside these things you can actually hear the noise of the amplified chip and they want to mask that. So if you liked the video, please give it a big thumbs up. And as always, comment down below Oh Over on the Eevblog for me, let me know if you want me to, you know, do anything further with this and to try and get it eliminated. But wait yeah we can get it down a little bit.

but I don't think we are going to be able to get at the zero unless we change that amplifier chip and you can't just get I've looked and you can't just get like a pin compatible or even a pin similar compatible one that you know you could bog in or something like that. They all have this inherent noise floor Anyway, catch you next time.