Thats interesting....
I wrote something a long time ago in this topic.
Then I decided to make a PGA2311 preamp now! With a friend who can write some PIC codes...
Anyway,
I dont know if error401 already made this preamp or not. I'd like to ask something to be sure.
- This IC has (PGA2311) a really good channel seperating value (-130dB). So if we use a dual opamp as the input buffer then can we catch the same value on channel speration? Or should we use seperated opamps like OPA134 per channel?
- This IC has internal opamps, we know that. However there is no slew rate information for these opamps.. So what kind of opamps in the buffer give the best result by the meaning of total slew rate of that preamplifier?
- Somebody mentioned (I cannot remember who he is) before; if you use PGA without gain (<= 0dB) then you cannot hear anything while changing the volume level. However if you use PGA with some gains (internal gain) then you may hear some noise while changing the level.. Is this information true? Has anybody tried that?
- And a last question about the buffer again; some other body (that I also cannot remember his name) says the buffer with gain will sound better than an unity gain buffer.. Is also that true?
Thanks in advance for the answers. And also some additional informations except these answers...
I wrote something a long time ago in this topic.
Then I decided to make a PGA2311 preamp now! With a friend who can write some PIC codes...
Anyway,
I dont know if error401 already made this preamp or not. I'd like to ask something to be sure.
- This IC has (PGA2311) a really good channel seperating value (-130dB). So if we use a dual opamp as the input buffer then can we catch the same value on channel speration? Or should we use seperated opamps like OPA134 per channel?
- This IC has internal opamps, we know that. However there is no slew rate information for these opamps.. So what kind of opamps in the buffer give the best result by the meaning of total slew rate of that preamplifier?
- Somebody mentioned (I cannot remember who he is) before; if you use PGA without gain (<= 0dB) then you cannot hear anything while changing the volume level. However if you use PGA with some gains (internal gain) then you may hear some noise while changing the level.. Is this information true? Has anybody tried that?
- And a last question about the buffer again; some other body (that I also cannot remember his name) says the buffer with gain will sound better than an unity gain buffer.. Is also that true?
Thanks in advance for the answers. And also some additional informations except these answers...
Dxvideo said:Thats interesting....
I wrote something a long time ago in this topic.
Then I decided to make a PGA2311 preamp now! With a friend who can write some PIC codes...
Anyway,
I dont know if error401 already made this preamp or not. I'd like to ask something to be sure.
- This IC has (PGA2311) a really good channel seperating value (-130dB). So if we use a dual opamp as the input buffer then can we catch the same value on channel speration? Or should we use seperated opamps like OPA134 per channel?
- This IC has internal opamps, we know that. However there is no slew rate information for these opamps.. So what kind of opamps in the buffer give the best result by the meaning of total slew rate of that preamplifier?
- Somebody mentioned (I cannot remember who he is) before; if you use PGA without gain (<= 0dB) then you cannot hear anything while changing the volume level. However if you use PGA with some gains (internal gain) then you may hear some noise while changing the level.. Is this information true? Has anybody tried that?
- And a last question about the buffer again; some other body (that I also cannot remember his name) says the buffer with gain will sound better than an unity gain buffer.. Is also that true?
Thanks in advance for the answers. And also some additional informations except these answers...
I remember hearing from someone somewhere that the internal output stage in the PGA23xx is based on OPA2132, but I'm not sure how reliable that is.
My plan is to use LME49720 as the input buffer, possibly with an active low-pass, and no output buffer; PGA2320 has an internal opamp that can drive any reasonable load, I don't see why another stage is necessary. This opamp has pretty impressive specs, and I think it sounds good, so I'm fine with it - but it's not particularly fast, and the channel separation isn't as good (118dB at 1KHz). I'm going to do a 6 channel preamp so I won't have room for single channel opamps, but you may choose to prioritize differently and avoid crosstalk - if you go for individual chips, maybe look at LMH6321 which is apparently very good (and it's extremely fast). Keep in mind that getting 130dB of channel separation is going to be very dependant on board layout as well; even in my design board layout probably has more effect than the ICs.
I haven't done lots of testing with the PGA's yet, but I don't think they should make noise when changing volume levels, gain or not. There is internal zero crossing detection that is optional (I don't see why you'd disable it) that should prevent this - but if the ZCD is not used, they will certainly make noise when switching settings. I don't think gain would really come into it. I haven't tested any of this though...
Whether the buffer sounds better with or without gain is going to be very opamp dependent. It likely has its roots in the gain compensation and instability at unity gain for most designs without compensation. I'm not sure I'd put much stock in it though, I have never heard this effect myself.
I've put this project on the back burner for now, and basically what I'm doing is prototyping each individual section before I take on the whole thing at once. So right now I'm working on a PGA2320 (or PGA2310) volume control board, there's a thread over at head-fi if you want to check it out - but no onboard buffers at all. I'm working on a dedicated website for these projects as well.
Good luck on your project
error401 said:
I remember hearing from someone somewhere that the internal output stage in the PGA23xx is based on OPA2132, but I'm not sure how reliable that is.
My plan is to use LME49720 as the input buffer, possibly with an active low-pass, and no output buffer; PGA2320 has an internal opamp that can drive any reasonable load, I don't see why another stage is necessary. This opamp has pretty impressive specs, and I think it sounds good, so I'm fine with it - but it's not particularly fast, and the channel separation isn't as good (118dB at 1KHz). I'm going to do a 6 channel preamp so I won't have room for single channel opamps, but you may choose to prioritize differently and avoid crosstalk - if you go for individual chips, maybe look at LMH6321 which is apparently very good (and it's extremely fast). Keep in mind that getting 130dB of channel separation is going to be very dependant on board layout as well; even in my design board layout probably has more effect than the ICs.
I haven't done lots of testing with the PGA's yet, but I don't think they should make noise when changing volume levels, gain or not. There is internal zero crossing detection that is optional (I don't see why you'd disable it) that should prevent this - but if the ZCD is not used, they will certainly make noise when switching settings. I don't think gain would really come into it. I haven't tested any of this though...
Whether the buffer sounds better with or without gain is going to be very opamp dependent. It likely has its roots in the gain compensation and instability at unity gain for most designs without compensation. I'm not sure I'd put much stock in it though, I have never heard this effect myself.
I've put this project on the back burner for now, and basically what I'm doing is prototyping each individual section before I take on the whole thing at once. So right now I'm working on a PGA2320 (or PGA2310) volume control board, there's a thread over at head-fi if you want to check it out - but no onboard buffers at all. I'm working on a dedicated website for these projects as well.
Good luck on your project
Thanks my friend,
The informations will be very useful for my project...
Were not started the project yet. Still working on purchasing!
Anyway,
I was planning to use one LM4562 or OPA2132 for the input buffer, may be two OPA134 or LT115 I can use for that position.
Also I dont want to use an output buffer. I am agree with you, if PGA can drive 600R loads then why should I use a buffer on outputs?
I think simplistic, so there will be just;
- one PIC16877
- two ULN2004 (for 8 input selecting, 2 power control, 1 amplifier mute and 3 spare relay output)
- one PGA2311PA (PA series have better THD figure I think)
- two (or one) good opamp for the input buffer.
- one 2 x 16 LCD
- one rotary encoder with push button
- one IR receiver
I expect to make all functions with that rotary encoder. We will see.
I will publish the results from this topic. I hope we will success..
By the way;
I plan to make not just a preamplifier. May be I can use that circuýit for an integrated amplifier with a gainclone. However in fact I was planning to make a buffered gainclone (Pedja Rogic's JFET buffered) But under that conditions, the buffer looks not necessary (if you take consider we already have a buffer before the volume control!) ... Do you agree with me?
howdy! sorry to resurrect this thread, but i've been searching on the PGA series of preamp chips lately...
Error401, if you're going to use your uC to ramp the volume down then up, you'll need to block DC at either the inputs of the PGA or the inputs of the buffer opamp, or both. any DC at the input of the PGA during switching, even if ZCEN is enabled, will cause some zipper noise, IIRC.
i'm not sure if you've completed this project yet, but i thought i'd post my two cents. i'm currently in the process of designing such a board around the PGA2310 (higher analog rails). the thread relating to the design (with great insight on IO/ground layout) is at this link.
also, on VFDs: digikey and mouser both carry newhaven display VFDs on the cheap (my design calls for a 2x16 that costs only $30)
if you're done with this, how'd it turn out? if not, good luck!
~ brad.
error401 said:
Error401, if you're going to use your uC to ramp the volume down then up, you'll need to block DC at either the inputs of the PGA or the inputs of the buffer opamp, or both. any DC at the input of the PGA during switching, even if ZCEN is enabled, will cause some zipper noise, IIRC.
i'm not sure if you've completed this project yet, but i thought i'd post my two cents. i'm currently in the process of designing such a board around the PGA2310 (higher analog rails). the thread relating to the design (with great insight on IO/ground layout) is at this link.
also, on VFDs: digikey and mouser both carry newhaven display VFDs on the cheap (my design calls for a 2x16 that costs only $30)
if you're done with this, how'd it turn out? if not, good luck!
~ brad.
The whole preamp idea I haven't had time for, but I did end up experimenting with the PGA2320 and building a carrier board/controller for it. That project is documented here if you're interested.
I'm not satisfied with character VFDs, they just look plain ugly to me. If it was a choice between using that and indicator LEDs alone, I'd just take the LEDs and skip the display. I want a graphic VFD!
I do get zipper noise occasionally while changing volume, but it seems to be more of a problem while manually manipulating the volume than it is with the ramping, which I haven't had any trouble with. Blocking DC is probably a good idea if/when I get to the integrated preamp though, thanks for the input.
I'm not satisfied with character VFDs, they just look plain ugly to me. If it was a choice between using that and indicator LEDs alone, I'd just take the LEDs and skip the display. I want a graphic VFD!
I do get zipper noise occasionally while changing volume, but it seems to be more of a problem while manually manipulating the volume than it is with the ramping, which I haven't had any trouble with. Blocking DC is probably a good idea if/when I get to the integrated preamp though, thanks for the input.
Hi, could figure out where else to post this, but I have a small problem with the PGA2310 I've been playing around with, the problem is there is no sound until the volume is half way, I have a LCD display just showing the step from 0-255 that is being sent to the PGA2310 so when it get to 125-130 there is sound all of a sudden
Thanks
Thanks
my source code is also displaying the binary on the lcd showing me whats being sent to the PGA in 16bits where both of the 8bits are mirrored to increase both channels at the same time, right now I have the mute pin connected to nothing, I never thought it would be a problem if it was floating because I didn't see anything in the datasheet about it, but i'll try that next.
Don't ever leave digital input pins floating. They tend to accumulate charge and you'll end up with some voltage there. Sometimes you get a '0' other times a '1', and most of the time somewhere in between that will cause metastability (if you're lucky) or latch-up and dead chips (if you're unlucky).
I'm not sure why you bother to ramp the volume control manually. Toggling the MUTE pin will do the same thing. The chip actually does ramp the volume on mute/un-mute.
It's interesting... I just completed my preamp that I have designed over the past couple of years and finally decided executing on. Input select via relays controlled by an ULN28xx driver, LME49720 buffer, 1 Hz high-pass (DC block), PGA2320 volume control, LME49720 output buffer. I'm using 7-segment LED displays (yes, I'm old school) to indicate the volume. The display is bright when input or volume is changed but dims to a discrete dark red after 2 seconds (at which point the uC goes to sleep). Everything is controlled by a ATmega165. I use encoders for volume control and input select along with an IR receiver. The LED dimming is done without resorting to PWM as I didn't want any noise radiators in the amp.
Beware that with the PGA chips, you need to turn all powers off when you power off your amp. In my amp, the digital +5V stays live when the amp is in stand-by. I spoke with TI's applications engineers about this. They were a bit concerned about latch-up if just the 5V was left on. So I'm switching it with a MOSFET.
Sound quality: I never realized how much difference a preamp could make. Seriously! I like it. It's open and accurate but without the aggressive hissy treble that I have heard from other amps. I have compared it with a friend's Parasound P3 (based on OPA2134, PGA2310). The Parasound is a solid preamp! But mine is way more laid back and less tense to listen to. My friend was actually tempted to swap the OPA2134's in the Parasound with LME49720. The less aggressive treble made metallic instruments like cymbals, hi-hat, etc. sound less hissy and more metalic.
One thing to keep in mind is that the PGA23xx parts are not super low noise. They're about 10 dB noisier than the LME49720 (that granted are state-of-the-art). But it is noisier than a 100 kOhm potentiometer. I am only able to measure the difference. In actual use, I can't tell a difference. I've attached the noise plots for academic interest. The plots show the noise floor before and after the volume control with the volume control set to 0.00 dB.
~Tom
I'm not sure why you bother to ramp the volume control manually. Toggling the MUTE pin will do the same thing. The chip actually does ramp the volume on mute/un-mute.
It's interesting... I just completed my preamp that I have designed over the past couple of years and finally decided executing on. Input select via relays controlled by an ULN28xx driver, LME49720 buffer, 1 Hz high-pass (DC block), PGA2320 volume control, LME49720 output buffer. I'm using 7-segment LED displays (yes, I'm old school) to indicate the volume. The display is bright when input or volume is changed but dims to a discrete dark red after 2 seconds (at which point the uC goes to sleep). Everything is controlled by a ATmega165. I use encoders for volume control and input select along with an IR receiver. The LED dimming is done without resorting to PWM as I didn't want any noise radiators in the amp.
Beware that with the PGA chips, you need to turn all powers off when you power off your amp. In my amp, the digital +5V stays live when the amp is in stand-by. I spoke with TI's applications engineers about this. They were a bit concerned about latch-up if just the 5V was left on. So I'm switching it with a MOSFET.
Sound quality: I never realized how much difference a preamp could make. Seriously! I like it. It's open and accurate but without the aggressive hissy treble that I have heard from other amps. I have compared it with a friend's Parasound P3 (based on OPA2134, PGA2310). The Parasound is a solid preamp! But mine is way more laid back and less tense to listen to. My friend was actually tempted to swap the OPA2134's in the Parasound with LME49720. The less aggressive treble made metallic instruments like cymbals, hi-hat, etc. sound less hissy and more metalic.
One thing to keep in mind is that the PGA23xx parts are not super low noise. They're about 10 dB noisier than the LME49720 (that granted are state-of-the-art). But it is noisier than a 100 kOhm potentiometer. I am only able to measure the difference. In actual use, I can't tell a difference. I've attached the noise plots for academic interest. The plots show the noise floor before and after the volume control with the volume control set to 0.00 dB.
~Tom
Attachments
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.