• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Help with odd problem: SMSL AO200 emits hissing sound only when connected to a tweeter

RandomEar

Active Member
Joined
Feb 14, 2022
Messages
178
Likes
281
OK, so this is an odd problem: I've recently bought an SMSL AO200 as a stop-gap solution, until I find a Hypex or Purify implementation I like. Now, I've had two units of the AO200 already and both exhibit a peculiar behaviour: When the amp is connected to at least one speaker and is not muted, the amp and the speaker emit a very high pitched, low volume hissing sound. It sounds sharp and also has some crackling character. Spectrum analysis is inconclusive because the noise isn't very loud, but results indicate a possible mix of 4/8/12 kHz tones + some other stuff.

I've now debugged this noise and found the following: The hissing/crackling is only emitted, when the amp has a connection to the tweeters of my ELAC 243. The speakers can be bi-amped and I tested it with only the base driver connected, which results in the amp being silent. Nothing else I did (different wall outlet, disconnecting/turning off all other devices in the room, connecting only the cable without speakers, re-routing the cable) changed anything in the outcome.

My conculsion is, that the amp is somehow "incompatible" with the frequency divider for the tweeter in my ELACs and seems to oscillate when connected to that. Did anybody else ever experience anything like this? Any tips on how to possibly fix this? :confused:
 

dougi

Addicted to Fun and Learning
Joined
May 31, 2020
Messages
675
Likes
561
Location
ACT, Australia
Well the noise appears fairly high in frequency, so logical that it is only heard from the tweeter, so not a surprise. Is there any input connected (e.g. perhaps USB) when this problem occurs? It doesn't seem from the odd subjective review that this noise is normal. Have you got any other speakers lying around that you can also try?
 
OP
RandomEar

RandomEar

Active Member
Joined
Feb 14, 2022
Messages
178
Likes
281
Well the noise appears fairly high in frequency, so logical that it is only heard from the tweeter, so not a surprise. Is there any input connected (e.g. perhaps USB) when this problem occurs? It doesn't seem from the odd subjective review that this noise is normal. Have you got any other speakers lying around that you can also try?
Tested three paths: Either nothing connected other than the power cord, or connected over USB to PC or connected to DAC by XLR. Makes no difference.

The noise also comes from the amp itself. Therefore, even when only the base drivers are connected, the noise should still be audible at the amp. But it isn't.

Sadly, I don't have more speakers at hand, right now. I'll see what I can come up with.
 
OP
RandomEar

RandomEar

Active Member
Joined
Feb 14, 2022
Messages
178
Likes
281
I tested it with speakers wired in reversed polarity today. That also didn't make a difference. Neither did any settings on the amp. I also checked the spectrum again: On the speakers, there's a dominant 200 Hz tone with some 100 & 50 Hz too - looks like mains noise. The high pitched stuff is difficult to measure at the speakers, but is present. On the amp, the high-pitched noise is easier to measure and shows 4, 6 and 8 kHz tones, plus likely some stuff around 10.2 kHz or so and maybe one or two more tones higher up. But the ones above 8 kHz are too low in volume to be sure.

Didn't get any other speakers in to test, yet. I may just get rid of the amp - I'll see how motivated I am to troubleshoot this further.
 

norcalscott

Active Member
Forum Donor
Joined
Oct 30, 2020
Messages
170
Likes
254
Location
Sierra Foothills
Get rid of the SMSL amplifier.
Curious - what do you think is wrong with SMSL amplifiers? I have an AO200 and have been quite pleased with it for the price. Mine is dead silent with gain at maximum and no input. I'm driving it balanced from an SMSL DAC balanced output.
 

dougi

Addicted to Fun and Learning
Joined
May 31, 2020
Messages
675
Likes
561
Location
ACT, Australia
It seems strange that both your samples have had this "issue" but not one that has been reported elsewhere. As you say, also the noise is also from the amp itself, but only when connected to the tweeters. Perhaps it doesn't like the impedance response of them or some other interaction if the chips are set to "filterless" mode. The only thing I can suggest is the same, swap out amp and speakers to try it.
 
OP
RandomEar

RandomEar

Active Member
Joined
Feb 14, 2022
Messages
178
Likes
281
Curious - what do you think is wrong with SMSL amplifiers? I have an AO200 and have been quite pleased with it for the price. Mine is dead silent with gain at maximum and no input. I'm driving it balanced from an SMSL DAC balanced output.
I suppose it's forming an oscillating curcuit when combined with the crossover of my tweeters. But that's speculation, of course.

It seems strange that both your samples have had this "issue" but not one that has been reported elsewhere. As you say, also the noise is also from the amp itself, but only when connected to the tweeters. Perhaps it doesn't like the impedance response of them or some other interaction if the chips are set to "filterless" mode. The only thing I can suggest is the same, swap out amp and speakers to try it.
Well, if nobody around here except me combined this amp with ELAC 243's and only those create the issue, I suppose it's logical to expect that nobody else reported this ;)

Just as an addendum, I tried different cables yesterday - shorter and lower cross-section. As expected, that made no difference. As I'm not really interested in getting new speakers, I guess the amp will have to go. Too bad :confused:
 

norcalscott

Active Member
Forum Donor
Joined
Oct 30, 2020
Messages
170
Likes
254
Location
Sierra Foothills
I suppose it's forming an oscillating curcuit when combined with the crossover of my tweeters. But that's speculation, of course.


Well, if nobody around here except me combined this amp with ELAC 243's and only those create the issue, I suppose it's logical to expect that nobody else reported this ;)

Just as an addendum, I tried different cables yesterday - shorter and lower cross-section. As expected, that made no difference. As I'm not really interested in getting new speakers, I guess the amp will have to go. Too bad :confused:
Yes, I would say that if two of these devices are exhibiting this problem it must be related to the combination. Probably no other remedy than to switch to a different amp. What about the new Topping LA90 or PA5 (although that model has had some issues)?
 
OP
RandomEar

RandomEar

Active Member
Joined
Feb 14, 2022
Messages
178
Likes
281
Yes, I would say that if two of these devices are exhibiting this problem it must be related to the combination. Probably no other remedy than to switch to a different amp. What about the new Topping LA90 or PA5 (although that model has had some issues)?
Due the known reliability concerns, I wouldn't consider the PA5. I also don't like the TRS connectors on that model. The LA90 is expensive and - due to being class A/B - inefficient. I'm really just looking for a cheap-ish class D stop-gap solution, until I find a suitable Hypex/Purifi build. I may also wait until Hypex brings a new NCx-version of their NC122MP or NC252MP to market and see how that performs. For now, I ordered an Aiyima A07 to check it out. Hope it's "good enough" for a couple of months.
 

Nonick

Active Member
Joined
Mar 27, 2021
Messages
108
Likes
50
I had the same problem with SMSL AO200 (hissing from tweeters). So i thought - ok, just a hiss to live with. Accidentaly i found it also had a problem with abnormal distorsion noise from volume 48+.
When unbalanced input was selected in AO200, nothing connected to unbalanced (only speaker output), it produced this noise:

Returned AO200 immediately and replaced it with PA5. No fancy display, no remote, but dead silent without hiss.
 
Last edited:

norcalscott

Active Member
Forum Donor
Joined
Oct 30, 2020
Messages
170
Likes
254
Location
Sierra Foothills
Due the known reliability concerns, I wouldn't consider the PA5. I also don't like the TRS connectors on that model. The LA90 is expensive and - due to being class A/B - inefficient. I'm really just looking for a cheap-ish class D stop-gap solution, until I find a suitable Hypex/Purifi build. I may also wait until Hypex brings a new NCx-version of their NC122MP or NC252MP to market and see how that performs. For now, I ordered an Aiyima A07 to check it out. Hope it's "good enough" for a couple of months.
I’m curious to get your impressions of the Aiyana - I have been considering that for a garage stereo system using an old pair of Tannoys. I’m using a similar device (Nobsound NS-03G) to replace a bad amp plate on a subwoofer and paired with an old Dell laptop power supply it sounds better than the original plate amp, and has way more power.
 
OP
RandomEar

RandomEar

Active Member
Joined
Feb 14, 2022
Messages
178
Likes
281
I’m curious to get your impressions of the Aiyana - I have been considering that for a garage stereo system using an old pair of Tannoys. I’m using a similar device (Nobsound NS-03G) to replace a bad amp plate on a subwoofer and paired with an old Dell laptop power supply it sounds better than the original plate amp, and has way more power.
Impressions of the A07 posted here. In short: It's small, light and cheap(ly build). It works. On the one hand, it's impressive what you get for the money. On the other hand, it doesn't seem to be quite what I'm looking for.
 
OP
RandomEar

RandomEar

Active Member
Joined
Feb 14, 2022
Messages
178
Likes
281
OK, so it's been a couple of weeks and I got ready to sell the AO200, but this problem has bugged me. I then stumbled upon this post by @arkelectron (+ follow ups), suggesting that I'm not the only one with that hiss & crackle noise. There are also people with an SMSL DA-9 (same Infineon chip) reporting a similar noise. A couple of days after reading those posts, @dr_mick51 by channce posted this observation about the filter design and expected speaker capacitance on amps with an Infineon MA12070 chip.

As far as I understand it, the situation is as follows: The MA12070 is designed to be run without chokes at the speaker outlets, as the designers at Infineon expect the speakers themselves to have sufficient capacitance and act as chokes, filtering out most HF noise. In the datasheet, the eval board of the MA12070 is measured with a 4 Ω load in series with a coil of 22 µH capacitance to simulate that of a "real" speaker. Some amp manufacturers seem to incorporate chokes in their MA12070-based designs. Some others, like SMSL in their A100 and most likely also their DA-9 and AO200, do not have those chokes installed. It may be a design choice, but it could also be a cost saving measure - I don't know.

At that point, I had the idea that the tweeters on my speakers could lack the capacitance to act as sufficient filters. And as a shot in the dark, I just got myself a couple of 22 µH, 3 A ferrite core chokes for 5 $. Put them in series with the tweeters, connected the amp and - silence. It works! A couple of measurements using my phone and the app "Spectroid":

Background spectrum
(Phone on the amp, amp off)

Screenshot_20220518-153656_BG on amp.png


Spectrum without chokes
(Phone on the amp, amp on)
Screenshot_20220518-153738_on amp, no choke.png


Spectrum with chokes
(Phone on the amp, amp on)
Screenshot_20220518-163534_on amp, choke.png


Yellow is the current level, red is a long term upper hold value. There's no crackle and hiss coming from the amp anymore. The noise from the speakers is also gone. Just the regular, very low level white noise is audible with my ear on the tweeter. Success!

"But what about changes in the frequency response or distortion?", you might ask. Well... I'm not 100% sure. My ears say that this sounds absolutely fine and there is no difference to the setup without chokes. Except for the lack of the crackle & hiss, of course. I also measured some test tones (1, 3, 5, 7, 11, 13, 15, 17 and 19 kHz) in front of the speaker with the same mobile app as shown above with and without chokes and they appear to be differing in level. Some tones above 10 kHz changed between 1 to 4 dB, but not in a uniform direction. Some are louder, some are quieter. And they also change inbetween measurements, which makes me suspect that the FFT resolution or the plot resolution of the app just aren't high enough to resolve those peaks properly.

For now, I'll keep the system running with chokes installed and see how it behaves. Maybe I get the option to accurately measure the difference some day. It goes without saying that, as a consumer, I should not have to fiddle around with random parts to get my amp to "agree" to my passive speakers. I got lucky on this one.
 

norcalscott

Active Member
Forum Donor
Joined
Oct 30, 2020
Messages
170
Likes
254
Location
Sierra Foothills
OK, so it's been a couple of weeks and I got ready to sell the AO200, but this problem has bugged me. I then stumbled upon this post by @arkelectron (+ follow ups), suggesting that I'm not the only one with that hiss & crackle noise. There are also people with an SMSL DA-9 (same Infineon chip) reporting a similar noise. A couple of days after reading those posts, @dr_mick51 by channce posted this observation about the filter design and expected speaker capacitance on amps with an Infineon MA12070 chip.
Very interesting and could explain why many do not experience this hissing (including me). I agree that you should not have to do these kinds of experiments, but I wonder if SMSL tested a number of speakers and none of the models they tested exhibited this? @SMSL-Mandy this may be good feedback to provide to support.
 

Toku

Major Contributor
Joined
Apr 4, 2018
Messages
1,836
Likes
1,868
Location
Japan
I use two AO200s, but I don't have such a symptom. If you watch the video that recorded the noise, I think that the cause is in the speaker circuit. Is there any place where the minus side is connected in common somewhere? Bi-amp drive cannot be used unless the speaker network is isolated in each band.
 

dr_mick51

Active Member
Joined
Mar 12, 2022
Messages
103
Likes
79
OK, so it's been a couple of weeks and I got ready to sell the AO200, but this problem has bugged me. I then stumbled upon this post by @arkelectron (+ follow ups), suggesting that I'm not the only one with that hiss & crackle noise. There are also people with an SMSL DA-9 (same Infineon chip) reporting a similar noise. A couple of days after reading those posts, @dr_mick51 by channce posted this observation about the filter design and expected speaker capacitance on amps with an Infineon MA12070 chip.

As far as I understand it, the situation is as follows: The MA12070 is designed to be run without chokes at the speaker outlets, as the designers at Infineon expect the speakers themselves to have sufficient capacitance and act as chokes, filtering out most HF noise. In the datasheet, the eval board of the MA12070 is measured with a 4 Ω load in series with a coil of 22 µH capacitance to simulate that of a "real" speaker. Some amp manufacturers seem to incorporate chokes in their MA12070-based designs. Some others, like SMSL in their A100 and most likely also their DA-9 and AO200, do not have those chokes installed. It may be a design choice, but it could also be a cost saving measure - I don't know.

At that point, I had the idea that the tweeters on my speakers could lack the capacitance to act as sufficient filters. And as a shot in the dark, I just got myself a couple of 22 µH, 3 A ferrite core chokes for 5 $. Put them in series with the tweeters, connected the amp and - silence. It works! A couple of measurements using my phone and the app "Spectroid":

Background spectrum
(Phone on the amp, amp off)

View attachment 207398

Spectrum without chokes
(Phone on the amp, amp on)
View attachment 207399

Spectrum with chokes
(Phone on the amp, amp on)
View attachment 207400

Yellow is the current level, red is a long term upper hold value. There's no crackle and hiss coming from the amp anymore. The noise from the speakers is also gone. Just the regular, very low level white noise is audible with my ear on the tweeter. Success!

"But what about changes in the frequency response or distortion?", you might ask. Well... I'm not 100% sure. My ears say that this sounds absolutely fine and there is no difference to the setup without chokes. Except for the lack of the crackle & hiss, of course. I also measured some test tones (1, 3, 5, 7, 11, 13, 15, 17 and 19 kHz) in front of the speaker with the same mobile app as shown above with and without chokes and they appear to be differing in level. Some tones above 10 kHz changed between 1 to 4 dB, but not in a uniform direction. Some are louder, some are quieter. And they also change inbetween measurements, which makes me suspect that the FFT resolution or the plot resolution of the app just aren't high enough to resolve those peaks properly.

For now, I'll keep the system running with chokes installed and see how it behaves. Maybe I get the option to accurately measure the difference some day. It goes without saying that, as a consumer, I should not have to fiddle around with random parts to get my amp to "agree" to my passive speakers. I got lucky on this one.
I'm glad you were able to find the cause of your issues. But the way, what chokes did you buy?
 
OP
RandomEar

RandomEar

Active Member
Joined
Feb 14, 2022
Messages
178
Likes
281
I use two AO200s, but I don't have such a symptom. If you watch the video that recorded the noise, I think that the cause is in the speaker circuit. Is there any place where the minus side is connected in common somewhere? Bi-amp drive cannot be used unless the speaker network is isolated in each band.
Yes, the cause is the tweeter crossover. The connection on my speakers is a bridge you can take out, which connects (+) and (-) separately. The bridge is now replaced by one cable from woofer(-) to tweeter(-) and one choke connecting woofer(+) to tweeter (+). The amp cables are coming in at the woofer terminals.

I'm glad you were able to find the cause of your issues. But the way, what chokes did you buy?
I've got these.
 

dr_mick51

Active Member
Joined
Mar 12, 2022
Messages
103
Likes
79
OK, so this is an odd problem: I've recently bought an SMSL AO200 as a stop-gap solution, until I find a Hypex or Purify implementation I like. Now, I've had two units of the AO200 already and both exhibit a peculiar behaviour: When the amp is connected to at least one speaker and is not muted, the amp and the speaker emit a very high pitched, low volume hissing sound. It sounds sharp and also has some crackling character. Spectrum analysis is inconclusive because the noise isn't very loud, but results indicate a possible mix of 4/8/12 kHz tones + some other stuff.

I've now debugged this noise and found the following: The hissing/crackling is only emitted, when the amp has a connection to the tweeters of my ELAC 243. The speakers can be bi-amped and I tested it with only the base driver connected, which results in the amp being silent. Nothing else I did (different wall outlet, disconnecting/turning off all other devices in the room, connecting only the cable without speakers, re-routing the cable) changed anything in the outcome.

My conculsion is, that the amp is somehow "incompatible" with the frequency divider for the tweeter in my ELACs and seems to oscillate when connected to that. Did anybody else ever experience anything like this? Any tips on how to possibly fix this? :confused:ain, I'm sorry I did not read the whole thread since the beginning

OK, so this is an odd problem: I've recently bought an SMSL AO200 as a stop-gap solution, until I find a Hypex or Purify implementation I like. Now, I've had two units of the AO200 already and both exhibit a peculiar behaviour: When the amp is connected to at least one speaker and is not muted, the amp and the speaker emit a very high pitched, low volume hissing sound. It sounds sharp and also has some crackling character. Spectrum analysis is inconclusive because the noise isn't very loud, but results indicate a possible mix of 4/8/12 kHz tones + some other stuff.

I've now debugged this noise and found the following: The hissing/crackling is only emitted, when the amp has a connection to the tweeters of my ELAC 243. The speakers can be bi-amped and I tested it with only the base driver connected, which results in the amp being silent. Nothing else I did (different wall outlet, disconnecting/turning off all other devices in the room, connecting only the cable without speakers, re-routing the cable) changed anything in the outcome.

My conculsion is, that the amp is somehow "incompatible" with the frequency divider for the tweeter in my ELACs and seems to oscillate when connected to that. Did anybody else ever experience anything like this? Any tips on how to possibly fix this? :confused:
Man, I'm sorry I did not read all these since the beginning. So you were trying to use this amp to bi-amp a pair of speakers. Yes, when you connect directly to the tweeter terminals there are not enough "inductance", is the term. Because of this the amp does not have the right filtering. Adding the 22uH chokes before the tweeter clearly solves the problem.
But if you use the speakers with jumped terminals(not bi-amped) you should not have this issue because: one, the woofer has enough inductance per se, and two, it is very likely that the crossover has a big inductor in series with the woofer.
Now, I want you to think about what happened with Amir's measurements for the SMSL A100. He did not used those extra chokes, and you just confirmed the big difference it makes by using them. I would like to see how these Merus amplifiers measure when using the right filters. He agreed that those high frequency distortions were holding back the amp badly. Without them we could be talking about an almost 90 db SINAD. I don't remember any other Merus amp measured by Amir except the Loxjie A30 which has those 4 SMD chokes on the back of the board, and of course its measurements were better than A100.
Edited: I clearly forgot about the dual Merus Topping PA3s which measured very well. I would bet it also has these 4 chokes somewhere. I have not seen the back of the board pictures.
 
Last edited:
OP
RandomEar

RandomEar

Active Member
Joined
Feb 14, 2022
Messages
178
Likes
281
Man, I'm sorry I did not read all these since the beginning. So you were trying to use this amp to bi-amp a pair of speakers. Yes, when you connect directly to the tweeter terminals there are not enough "inductance", is the term. Because of this the amp does not have the right filtering. Adding the 22uH chokes before the tweeter clearly solves the problem.
But if you use the speakers with jumped terminals(not bi-amped) you should not have this issue because: one, the woofer has enough inductance per se, and two, it is very likely that the crossover has a big inductor in series with the woofer.
Now, I want you to think about what happened with Amir's measurements for the SMSL A100. He did not used those extra chokes, and you just confirmed the big difference it makes by using them. I would like to see how these Merus amplifiers measure when using the right filters. He agreed that those high frequency distortions were holding back the amp badly. Without them we could be talking about an almost 90 db SINAD. I don't remember any other Merus amp measured by Amir except the Loxjie A30 which has those 4 SMD chokes on the back of the board, and of course its measurements were better than A100.
Edited: I clearly forgot about the dual Merus Topping PA3s which measured very well. I would bet it also has these 4 chokes somewhere. I have not seen the back of the board pictures.
No, you were right before and got it mixed up now ;) The speakers can be bi-amped, but I never did that. I just tested the woofer and tweeter individually (bridge removed) for diagnostic purposes. The problem occurs, as soon as the tweeters are connected to the amp in any form: The amp whines, when the bridge is in place and the amp is connected to the speakers and driving both, tweeters and woofers in parallel. It also whined, when I connected only the tweeters for testing, without the bridge. It did not whine when I connected only the woofers without the bridge, also for testing.

As soon as I installed the choke on both speakers, everything was solved. I tested the choke in series with the whole speaker (amp(+) -> choke -> woofer(+) -> bridge -> tweeter(+)) and only in series with the tweeters (as described in post #18: amp(+) -> woofer(+) -> choke -> tweeter(+), woofer-tweeter bridge only on (-)). Both solutions remedied the hiss. This also sounds plausible to me, as without a choke, any HF noise would pass right through the low inductance tweeters, regardless of the woofer inductance. But my EE-knowledge is limited, so maybe I'm on the wrong track with that interpretation.

Anyway, your post got me thinking and lead to the solution, so thanks a lot! :)
 
Top Bottom