Thanks for doing the maths. I think it may be even less than that in practice. NCore is 400w into 4 ohm, and it's likely that the 70w is into 8 ohm (and more into 4 ohm)
While we are on the subject I have a (noob) question: is the R141 part of the gain loop of the amplifier, or just an attenuator in front of it?
In other word, does removing this resistor and lowering the gain by ~13dB also reduce the residual noise of the amplifier when no source is connected ?
In other word, does removing this resistor and lowering the gain by ~13dB also reduce the residual noise of the amplifier when no source is connected ?
Why not just give me a clear answer? 🙄
I understand the "do your homework" thinking on a diy forum, but on the other hand one cannot have a clear understanding of everything in this hobby, and I am admittedly a noob when it comes to electronic, so please bear with me...
I understand the "do your homework" thinking on a diy forum, but on the other hand one cannot have a clear understanding of everything in this hobby, and I am admittedly a noob when it comes to electronic, so please bear with me...
Thanks Julf
So I can expect the residual noise of the amp itself (input shorted) to go from 23uV down to under 6uV ? That is impressive.
So I can expect the residual noise of the amp itself (input shorted) to go from 23uV down to under 6uV ? That is impressive.
It's part of the feedback loop of a separate input buffer stage. There is no overall feedback from the output of the amplifier to the inputs of the buffer stage. (This can be seen from the application note)
It depends which part of the complete circuit is the limiting factor for the noise floor. If it's the input stage then lowering the gain may lower the noise floor. If it's the class D stage then lowering the gain will have no effect on the noise floor. Unless someone has already done this and made measurements, then apart from those at Hypex I doubt anyone in this thread knows the answer.
There is no clear answer.
Or, the application note provides a clear answer to the first part of your question ... if you know how to interpret it 🙂
It is always better to understand any answer in the context of the device being examined. The pointer link was definitely intended to help. It was not clear from your original question if this was all the level of information you required.
It depends which part of the complete circuit is the limiting factor for the noise floor. If it's the input stage then lowering the gain may lower the noise floor. If it's the class D stage then lowering the gain will have no effect on the noise floor. Unless someone has already done this and made measurements, then apart from those at Hypex I doubt anyone in this thread knows the answer.
There is no clear answer.
Or, the application note provides a clear answer to the first part of your question ... if you know how to interpret it 🙂
It is always better to understand any answer in the context of the device being examined. The pointer link was definitely intended to help. It was not clear from your original question if this was all the level of information you required.
Thanks ChrisPa
The NC1200 datasheet states 20uV of residual noise for a 27.8dB gain using the standard buffer, and 7uV for an 11.6dB gain unbuffered.
I don't know if this can be translated to the NC400, but that would indicate that some noise also comes from the classD stage.
That's the tricky part for me 😀
The NC1200 datasheet states 20uV of residual noise for a 27.8dB gain using the standard buffer, and 7uV for an 11.6dB gain unbuffered.
I don't know if this can be translated to the NC400, but that would indicate that some noise also comes from the classD stage.
I think the reasonable answer is if you have to reduce the gain, removing (or increasing) R141 is better than an attenuator. There is no reason to touch R141 unless you need to reduce the gain.
That being said, the difference is so small it doesn't matter in practice. The NCore itself already have lower distortion than a speaker so a few dB more feedback and lower distortion won't be noticable.
If you use attenuator before then you won't have to worry about ruining the NCores which is nice =)
I agree - as I keep saying, removing or changing R141 is not for someone with clumsy hands. I am pretty sure the difference between lowering gain and using an attenuator is totally inaudible in practice.
I can hear noise coming from the amp (no source connected) when hooked to a compression driver.
It looks like the only way to reduce this, beside using an lpad or autotransformer between the amp and driver, is to alter R141 and reduce any noise coming from the input stage.
It looks like the only way to reduce this, beside using an lpad or autotransformer between the amp and driver, is to alter R141 and reduce any noise coming from the input stage.
At 400-600 W / channel? I think they would be seriously battery-limited, unless you use the whole battery bank from a Tesla S... 🙂
- Status
- Not open for further replies.