Hi,
I have been building an preamplifier. Volume controller is PGA2310, but it has rather large
output noise, which is irratating(just hearable within 0.5m from speaker) at low volumes. Chip is between two operational amplifiers(OPA 227)
Then I got a "brigth" idea, to lower the noise of chip. Before the chip the voltage levels are
raised to four times original, and after the chip the voltage levels are lowered to line level signal(divided by four), so the noise should be also divided by four, am I correct?
At least the output noise is lowered to point which it is hearable about 10 cm from speaker.
Now was thinking maybe the output level could be divided futhermore, at least 3db more. But there is an possibility that the volume levels aren't suffient anymore, if I want someday use my system with less sensitive speakers(now sensitivity is 89db/1W) the volume levels aren't suffient anymore.
Is there any other way to reduse the noise on chip?
I have been building an preamplifier. Volume controller is PGA2310, but it has rather large
output noise, which is irratating(just hearable within 0.5m from speaker) at low volumes. Chip is between two operational amplifiers(OPA 227)
Then I got a "brigth" idea, to lower the noise of chip. Before the chip the voltage levels are
raised to four times original, and after the chip the voltage levels are lowered to line level signal(divided by four), so the noise should be also divided by four, am I correct?
At least the output noise is lowered to point which it is hearable about 10 cm from speaker.
Now was thinking maybe the output level could be divided futhermore, at least 3db more. But there is an possibility that the volume levels aren't suffient anymore, if I want someday use my system with less sensitive speakers(now sensitivity is 89db/1W) the volume levels aren't suffient anymore.
Is there any other way to reduse the noise on chip?
Hi fallow ,
Are you sure ? The noise from PGA2310 chip .
These chip PGA2310 , CS3310 is good in volume control .
This site ( http://smartlee.idv.tw/pic/cs3310report.htm) have A.P.
test report use cs3310 chip .
Check your layout or power supply (Va & Vd is isolate) ?
good luck !
Are you sure ? The noise from PGA2310 chip .
These chip PGA2310 , CS3310 is good in volume control .
This site ( http://smartlee.idv.tw/pic/cs3310report.htm) have A.P.
test report use cs3310 chip .
Check your layout or power supply (Va & Vd is isolate) ?
good luck !
Weird...
The only way I could have the pga's produce noise is to give them a setting above +15dB. Otherwise they are extremely quiet.
The only difficulty I have is some HF (25mV @ 10MHz) PSU noise on 'all' supplies which leaks through to the audio ouputs as well. But then again I seem to have trouble hearing that
You did connect the groundplanes for analog and digital grounds together ? Not that it matters a lot for the noise.
A8
The only way I could have the pga's produce noise is to give them a setting above +15dB. Otherwise they are extremely quiet.
The only difficulty I have is some HF (25mV @ 10MHz) PSU noise on 'all' supplies which leaks through to the audio ouputs as well. But then again I seem to have trouble hearing that
You did connect the groundplanes for analog and digital grounds together ? Not that it matters a lot for the noise.
A8
For the power supply, there are no picture available now,
but it is done by single 15VA 2x15V toroidial transformer. The voltage regulation to +5 V and +15 V is done by LM 317 and for -15V by LM337. Before each regulators there are two 1000uF capacitors(Panasonic FC) and one 100nF, after the regulator there are 3 220 nF and one 100 nF. The ground planes of digital and analog supply is compined rigth after diode bridge.(trying to find them somewhere if you wish to see them, and think they would be helpfull)
The output level of preamplifier, when no signal is feeded, is <20 micro voltages, don't know for sure, because the base noise on spectrum analyzer were around 20 micro voltages. (and when the amplifier is on or off the noise level did not rise.)
And when feeding a signal the signal peek is showing around amplification -92dB to -93dB.
Before dividing the output level by four(increasing before PGA2310 by four) when turning amplifier on rised the noise level in analyzer.
In both channel the behavioral is similar.
but it is done by single 15VA 2x15V toroidial transformer. The voltage regulation to +5 V and +15 V is done by LM 317 and for -15V by LM337. Before each regulators there are two 1000uF capacitors(Panasonic FC) and one 100nF, after the regulator there are 3 220 nF and one 100 nF. The ground planes of digital and analog supply is compined rigth after diode bridge.(trying to find them somewhere if you wish to see them, and think they would be helpfull)
The output level of preamplifier, when no signal is feeded, is <20 micro voltages, don't know for sure, because the base noise on spectrum analyzer were around 20 micro voltages. (and when the amplifier is on or off the noise level did not rise.)
And when feeding a signal the signal peek is showing around amplification -92dB to -93dB.
Before dividing the output level by four(increasing before PGA2310 by four) when turning amplifier on rised the noise level in analyzer.
In both channel the behavioral is similar.
Fallow
Did you try to connect the groundplanes together right next to (or under) the pga chip ?
This is stated in the product sheet.
Right now your 2 PSU's only share a common ground way back with the big caps, seems like a pretty long way for noise which can be picked up. I've found the 2310 to be quite sensitive to potential differences (and noise) between the digital and analog grounds. Problems went away when I connected them together next to the chip - minimal distance possible.
HTH
A8
Did you try to connect the groundplanes together right next to (or under) the pga chip ?
This is stated in the product sheet.
Right now your 2 PSU's only share a common ground way back with the big caps, seems like a pretty long way for noise which can be picked up. I've found the 2310 to be quite sensitive to potential differences (and noise) between the digital and analog grounds. Problems went away when I connected them together next to the chip - minimal distance possible.
HTH
A8
fallow ~ have you resolved the problem with noise? I would like to use PGA2310 in extremaly low noise system based on OPA1632. Unfortunately I can't go down with the noise of PGA2310 to meet the level of the rest of my audio path It is not real possible that so expensive chip has so big noise level ... isn't it?
It is not real possible that so expensive chip has so big noise level ... isn't it?
Having looked at ur schematics, they seem alright to me, however, only problem regarding noise in my point of view is digital and anlogue grounds, have a look at the link this is the design that was initiated and designed by myself and another friend it was completed few weeks back and its working perfectly no noise problems for us.
I have done couple of stereo versions as well using PIC or 8051 mcu but no problem with noise ever.
I would suggest to keep digital and analogue grounds seperate thats what i do myself "always", and for 5v digital supply trafo should have an independent winding.
http://www.diyaudio.com/forums/showthread.php?threadid=63786&highlight=
This PSU design of using single winding for +15v analogue and +5v digital is not recomended by at leat me.
I have done couple of stereo versions as well using PIC or 8051 mcu but no problem with noise ever.
I would suggest to keep digital and analogue grounds seperate thats what i do myself "always", and for 5v digital supply trafo should have an independent winding.
http://www.diyaudio.com/forums/showthread.php?threadid=63786&highlight=
This PSU design of using single winding for +15v analogue and +5v digital is not recomended by at leat me.
abidr ~ what mean "no problem with noise" ? The noise level I have in my system is really low, but ... it is still audible I have one question – If you switch on your audio system (volume set to max) and you place your ear 1cm from the tweeter – could you hear the noise? If the answer is ‘yes’ the noise level is to high for me I would like to have noise less system - it is possible using OPA1632 in fully balanced path.
I have one question – If you switch on your audio system (volume set to max) and you place your ear 1cm from the tweeter – could you hear the noise? If the answer is ‘yes’ the noise level is to high for me I would like to have noise less system - it is possible using OPA1632 in fully balanced path.
OK. – I made one more try I took just low noise Power supply stage MUR860->LM2941/2991+Rubycon ZL and single PGA2310. Inputs and outputs without any load – just simple supplied … and outputs I crossed to power stage input And I still have a very very low but from 1-10 cm distance from the tweeter audible noise from PGA2310. You can say it’s not important when it’s not audible from 0,5 meter distance from the speaker but … only the best possible systems are good enough So, I will try to ask directly TI guys about it – maybe exists a optimal application to minimize this noise – they should now it
I took just low noise Power supply stage MUR860->LM2941/2991+Rubycon ZL and single PGA2310. Inputs and outputs without any load – just simple supplied … and outputs I crossed to power stage input And I still have a very very low but from 1-10 cm distance from the tweeter audible noise from PGA2310. You can say it’s not important when it’s not audible from 0,5 meter distance from the speaker but … only the best possible systems are good enough So, I will try to ask directly TI guys about it – maybe exists a optimal application to minimize this noise – they should now it - Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.