Hissing from Genelec G Two at short range — should I be worried?

[Noklims]

Hi - I’ve just bought one Genelec G Two active monitor (one unit), but it has a slight hiss (white noise) even with all input cables disconnected.
The noise appears to come from the amplifier (no transformer/coils whining) and is mainly coming from the HF driver - it’s barely audible from the LF driver. In a very quiet room at night I can hear it up to about 40 cm away, in the daytime it’s only noticeable up to about 20 cm.

Would you say this is:

• completely normal
• a bit unlucky
• seriously unlucky
• defective - best to return it to the shop

 

[amirm]

It is very typical.

 

[Ilkka Rissanen]

Completely normal, all electronics/amplifiers produce some amount of noise which is transmitted through the drivers.

 

[Noklims]

Completely normal, all electronics/amplifiers produce some amount of noise which is transmitted through the drivers.

Got it, thanks.

By the way, are the Genelec G One, Genelec G Two and Genelec G Three roughly the same in terms of self-noise, or are some of them quieter due to differences in the amplifier design or the power supply filtering?

 

[Ilkka Rissanen]

Got it, thanks.

By the way, are the Genelec G One, Genelec G Two and Genelec G Three roughly the same in terms of self-noise, or are some of them quieter due to differences in the amplifier design or the power supply filtering?

They are very similar in terms of self noise, all specified at <5 dBA at 1 meter.

 

[-Matt-]

It seems to be expected that active speakers may have a slight hiss when you put your ear to them, but you don't generally notice this with passive speakers.

I think this is generally accepted because active speakers have other benefits, such as more control over the crossovers and built in dsp to correct resonances etc. (Also, this level of hiss is probably not audible at the listening distance, especially when masked by the music playing).

Just wondering, are there good (scientific) reasons why hiss is unavoidable in active speakers? At the end of the day, in both active and passive systems you have amplifiers and drivers, why is hiss accepted in one case but not the other?

 

[Ilkka Rissanen]

It seems to be expected that active speakers may have a slight hiss when you put your ear to them, but you don't generally notice this with passive speakers.

I think this is generally accepted because active speakers have other benefits, such as more control over the crossovers and built in dsp to correct resonances etc. (Also, this level of hiss is probably not audible at the listening distance, especially when masked by the music playing).

Just wondering, are there good (scientific) reasons why hiss is unavoidable in active speakers? At the end of the day, in both active and passive systems you have amplifiers and drivers, why is hiss accepted in one case but not the other?

Passive speakers/systems have hiss too, it is just at lower level and in many times not as audible. The difference comes from the fundamental difference in signal chain.

Active speaker: crossover filtering -> power amplier(s) -> drivers
Passive speaker: power amplier(s) -> crossover filtering -> drivers

In active speakers the power ampliers are directly connected to the drivers, there is nothing in between (in some cases there might be a single high pass capacitor on the tweeter output channel) attenuating the low level noise coming from the power ampliers. The amplifiers are seeing the raw response of the drivers without any attenuation which results in much higher acoustical sensitivity compared to a passive system which has been filtered in between the power amplifiers and the drivers. In many cases the difference can be up to 5 - 15 dB SPL / W depending on the raw response of the drivers and possible waveguides which boost the sensitivity even further. In passive speakers the crossover filtering acts up as a brake in between the power amplifiers and drivers, attenuating the low level noise considerably. The same filtering will also cause unlinearity and losses which are absent in an active speaker design.

 

[AnalogSteph]

They are very similar in terms of self noise, all specified at <5 dBA at 1 meter.

Which means that it's going to be progressively more significant the smaller the model, as listening distance tends to be reduced accordingly. People generally find 8030 hiss levels tolerable, 8020 so-so, and the 8010 might as well be qualifying as a princess on a pea (this model is particularly egregious as it has an extremely sensitive input and not every owner even knows about the -10dB DIP switch; I guess this design choice was geared towards level consistency in multichannel setups).

 

[Noklims]

People generally find 8030 hiss levels tolerable, 8020 so-so, and the 8010 might as well be qualifying as a princess on a pea (this model is particularly egregious as it has an extremely sensitive input and not every owner even knows about the -10dB.

Do you think this could be related to the physical size of the monitors? Larger cabinets allow for a larger PCB and potentially more space for additional filtering components in the power supply or amplifier stage. Could that make the larger models quieter in terms of self-noise?
The -10 dB switch does not reduce the noise - the hiss level stays exactly the same (at least on my Genelec G Two).

Active speaker: crossover filtering -> power amplier(s) -> drivers
Passive speaker: power amplier(s) -> crossover filtering -> drivers

Does that mean that if active monitors are described as having two amplifiers - one for the LF driver and one for the HF driver - they are more likely to have higher self-noise (because the crossover is placed before the amplifiers)?
And if a monitor is described as having a single amplifier for the whole speaker, does that imply the crossover is after the amplifier and therefore those monitors would likely have lower self-noise?
Could you clarify: both the Genelec G Two and the Genelec G Three are specified as having 25+25 W amplification - are the amplifier and power supply circuits in these two models completely identical, or are there differences between them?

 

[Mort]

In active speakers the power ampliers are directly connected to the drivers, there is nothing in between (in some cases there might be a single high pass capacitor on the tweeter output channel) attenuating the low level noise coming from the power ampliers. The amplifiers are seeing the raw response of the drivers without any attenuation which results in much higher acoustical sensitivity compared to a passive system which has been filtered in between the power amplifiers and the drivers. In many cases the difference can be up to 5 - 15 dB SPL / W depending on the raw response of the drivers and possible waveguides which boost the sensitivity even further. In passive speakers the crossover filtering acts up as a brake in between the power amplifiers and drivers, attenuating the low level noise considerably. The same filtering will also cause unlinearity and losses which are absent in an active speaker design.

This is fascinating. Thanks!

 

[amirm]

Does that mean that if active monitors are described as having two amplifiers - one for the LF driver and one for the HF driver - they are more likely to have higher self-noise (because the crossover is placed before the amplifiers)?

Only indirectly in that it means it has an active crossover before the amps. Otherwise it wouldn't need more than one amp.

The ones with one amp do indicate the classic crossover on the output of the amp meaning you just get convenience of a built-in amp, not the benefits of an active speaker system.

And if a monitor is described as having a single amplifier for the whole speaker, does that imply the crossover is after the amplifier and therefore those monitors would likely have lower self-noise?

They can although they can still have amps that have high noise floor. Even stand-alone systems with external amps can have tweeter hiss. This is the reason for getting a very quiet amplifier (one with great SNR as I measure). Indeed, many pass systems with external amps have tweeter noise if you put your ear next to them. It is just that studio monitors tend to be subjected to this a lot more.

 

[TimoJ]

If you search this site, there are many threads about Genelec hissing. Users say that Neumann monitors are not as noisy as Genelecs. So I guess they use better quality amplifiers or attenuate tweeter with passive components?
For example:

Do the Genelec 8030c hiss?

I am close to buying a pair of Genelec 8030c speakers for desktop listening at an arm lengths distance. I listened to them, and a bunch of other monitors in store, and really liked them. However I've read some posts stating that they hiss if you aren't playing anything on them, and that the...

www.audiosciencereview.com

Genelec 8030C too hissy - what to get instead - KH80, 120? Paired with Presonus sub8.

Genelec 8030c have a bit too loud self hiss for my 1 m near-field use, so looking for another solution. A bit limited in desk space so ideally not something a lot deeper than 20 cm. Got a cheap and cheerful Presonus Sub8 that high passes above 80 Hz to deal with the low frequencies. I see...

www.audiosciencereview.com

 

[Mort]

Users say that Neumann monitors are not as noisy as Genelecs.

Do they now...

I too would like to know if they are but I would only trust a competent measurement.

 

[Ilkka Rissanen]

Which means that it's going to be progressively more significant the smaller the model, as listening distance tends to be reduced accordingly. People generally find 8030 hiss levels tolerable, 8020 so-so, and the 8010 might as well be qualifying as a princess on a pea (this model is particularly egregious as it has an extremely sensitive input and not every owner even knows about the -10dB DIP switch; I guess this design choice was geared towards level consistency in multichannel setups).

Every 8000 series Genelec monitor has an identical input sensitivity when they left the factory. That is 86 dB SPL at -20 dBu signal for the analogue input. Most models have adjustment to lower it if required, either a potentiometer or a -10 dB DIP switch. SAM models can be adjusted via GLM.

G series models have 10 dB lower input sensitiviy compred to 8000 series. Every G series model has additional -10 dB DIP switch to further reduce the input sensitivity if required.

 

[Ilkka Rissanen]

Does that mean that if active monitors are described as having two amplifiers - one for the LF driver and one for the HF driver - they are more likely to have higher self-noise (because the crossover is placed before the amplifiers)?
And if a monitor is described as having a single amplifier for the whole speaker, does that imply the crossover is after the amplifier and therefore those monitors would likely have lower self-noise?
Could you clarify: both the Genelec G Two and the Genelec G Three are specified as having 25+25 W amplification - are the amplifier and power supply circuits in these two models completely identical, or are there differences between them?

No, the number of channels/ampliers in active speaker does not make any difference, it is about the fundamental difference between an active vs. passive design. A single amplifier channel can't drive more than a single driver if talking about an active design where the crossover filtering is ALWAYS done before the power amplification stage. Only passive speakers can be run with a single amplifier channel (per speaker) since the crossover filtering is done after the power amplification stage.

Both G Two and G Three have 50 + 50 W power amplifiers. They can be considered identical between the models.

 

[TimoJ]

Do they now...

I too would like to know if they are but I would only trust a competent measurement.

This thread had measurements, but looks like there is no data anymore.
edit: the data is still there, you just need to select Data page:

Hiss (S&R) - 2021 04 30

docs.google.com

Hiss List (S&R)

Lots of discussion about hiss lately, how perceptible or annoying it is, and if it matters. I've created a sheet with Sound & Recording measurements of hiss in active loudspeakers: https://docs.google.com/spreadsheets/d/1n62hGAdKptSSKC74_fIT_R7DxFJ60YR2_O9C8x-vYQI/edit?usp=sharing Viewers can...

www.audiosciencereview.com

 

[TimoJ]

 

[Noklims]

Every 8000 series Genelec monitor has an identical input sensitivity when they left the factory. That is 86 dB SPL at -20 dBu signal for the analogue input. Most models have adjustment to lower it if required, either a potentiometer or a -10 dB DIP switch. SAM models can be adjusted via GLM.

G series models have 10 dB lower input sensitiviy compred to 8000 series. Every G series model has additional -10 dB DIP switch to further reduce the input sensitivity if required.

Is my understanding correct that all Genelec G-series monitors produce 86 dB SPL when a −10 dBu signal is applied to the analogue input?

Also — on my G Two, engaging the −10 dB DIP attenuator does not change the measured noise floor (background noise). Why doesn't the noise level decrease when the input gain is reduced?

 

[Ilkka Rissanen]

Is my understanding correct that all Genelec G-series monitors produce 86 dB SPL when a −10 dBu signal is applied to the analogue input?

Correct.

Also — on my G Two, engaging the −10 dB DIP attenuator does not change the measured noise floor (background noise). Why doesn't the noise level decrease when the input gain is reduced?

Input sensitivity adjustement affects most effectively against the noise coming into the loudspeaker, say from any source device. It also has an effect into the self noise generated by the filter section of the loudspeaker itself. But it does not affect the self noise generated by power amplifier section because the input gain stage is located before the power amplifiers. Depending on model, the overall self noise can be dominated either by the filter or the power amplifier section, and depending on this, the input sensitivity adjustment may have smaller or larger effect on the overall self noise level.

 

[Ilkka Rissanen]

View attachment 511869

It is a bit shame that only 4 out of 10 Genelec monitors on that list are our current production models, all others are old and discontinued. Obviously, as technology evolves, self noise levels are coming down with every new model/platform. Our most recent 8380A is the "quietest" monitor we have ever produced, measuring under 0 dBA at 1 meter distance. Previous record holder holder was the 8331A.