I've got a chinese PGA2311 based preamp (New version PGA2311 3 ways Remote Pre-amplifier Preamp | eBay)
I know this might not be the most respected product design on these forums but to my ears (via active Dynaudio monitors) it sounds good and serves it's purpose.
Almost...
I posted another thread here about adding a line level "tape" output, a project which introduced the need to understand all the components in the preamp.
I already understand some of the basic functionality of the amp section, then there's the digital controller and LCD board, which I won't even consider, as long as it works.
What I still don't understand I have color marked in the attached schematic.
The power input caps - these varys with different eBay suppliers, but most are much higher (2200uF & 220uF) than TI recommendation in the PGA2311 datasheet (10uF & 0.1uF). What are the purpose of these? To supply instant current drain for audio peaks?
The audio input and output resistors In some designs these are accompanied by a X uF cap on each channel (to reduce input DC offset?). There are 220R input resistors and 330R output resistors. 220R at the input might be to raise input Z, but why would you want to increase output Z? And there are two 10k to ground - Why would you want to short circuit to ground?
Last, the Digital power supply of the PGA(?). This seems to be taken from the analog +5V, via 10R to pin 1 and 4 and on to ground via 10uF cap. In other designs this is taken from VCC and not analog +5V. What does the 10R and the 10 uF cap here? Are there any practical effects of supplying the digital with the "theoretical" analog power, than the "theoretical" digital power?
By "other designs" I'm mostly referring to this alternative schematic of the same chinese preamp: http://www.platenspeler.com/diy/preamps/remote/pga2311_schema.jpg
I know this might not be the most respected product design on these forums but to my ears (via active Dynaudio monitors) it sounds good and serves it's purpose.
Almost...
I posted another thread here about adding a line level "tape" output, a project which introduced the need to understand all the components in the preamp.
I already understand some of the basic functionality of the amp section, then there's the digital controller and LCD board, which I won't even consider, as long as it works.
What I still don't understand I have color marked in the attached schematic.
The power input caps - these varys with different eBay suppliers, but most are much higher (2200uF & 220uF) than TI recommendation in the PGA2311 datasheet (10uF & 0.1uF). What are the purpose of these? To supply instant current drain for audio peaks?
The audio input and output resistors In some designs these are accompanied by a X uF cap on each channel (to reduce input DC offset?). There are 220R input resistors and 330R output resistors. 220R at the input might be to raise input Z, but why would you want to increase output Z? And there are two 10k to ground - Why would you want to short circuit to ground?
Last, the Digital power supply of the PGA(?). This seems to be taken from the analog +5V, via 10R to pin 1 and 4 and on to ground via 10uF cap. In other designs this is taken from VCC and not analog +5V. What does the 10R and the 10 uF cap here? Are there any practical effects of supplying the digital with the "theoretical" analog power, than the "theoretical" digital power?
By "other designs" I'm mostly referring to this alternative schematic of the same chinese preamp: http://www.platenspeler.com/diy/preamps/remote/pga2311_schema.jpg
What I still don't understand I have color marked in the attached schematic.
The power input caps - these varys with different eBay suppliers, but most are much higher (2200uF & 220uF) than TI recommendation in the PGA2311 datasheet (10uF & 0.1uF). What are the purpose of these? To supply instant current drain for audio peaks?
The audio input and output resistors In some designs these are accompanied by a X uF cap on each channel (to reduce input DC offset?). There are 220R input resistors and 330R output resistors. 220R at the input might be to raise input Z, but why would you want to increase output Z? And there are two 10k to ground - Why would you want to short circuit to ground?
Last, the Digital power supply of the PGA(?). This seems to be taken from the analog +5V, via 10R to pin 1 and 4 and on to ground via 10uF cap. In other designs this is taken from VCC and not analog +5V. What does the 10R and the 10 uF cap here? Are there any practical effects of supplying the digital with the "theoretical" analog power, than the "theoretical" digital power?
By "other designs" I'm mostly referring to this alternative schematic of the same chinese preamp: http://www.platenspeler.com/diy/preamps/remote/pga2311_schema.jpg
View attachment 323144 [/QUOTE]
The power supply caps are often sized as 'this sounds about right' or 'this is what I had handy'. Sounds funny but its not as bad as it seems. The PGA draws so little power that anything above 1000uF or so gets you in diminishing returns.
Instant current drain? A signal of 5V into 10k requires 0.5mA - hardly a big current drain. The 10uF/0.1uF in the data sheet are for local decoupling, and assume that the power supply itself has been taken care of.
The 220ohms on the output side are required to protect the chip when the mute relays are engaged - they short the output to gnd. The 330 input are a (limited) protection of the input to external junk like people touching the signal input after having been sitting on nylon chairs...
A smallish (few 100) ohms on the output is a good idea to protect the preamp from accidental shorts and from too large cap loads (which might cuase instability). As a preamp designer, you have no control over whatever anybody hangs off of it, so you need to play better safe than sorry.
The 10 ohm to 10uF provides some additional smoothing for the supply to the digital supply pin. This is not necessary for the digital pin but might have been put in to avoid backtalk of digital hash into the main supply. Then again, it might have been put in because 'everybody does it' so it does look good. Don't look so surprised!
jan
The power input caps - these varys with different eBay suppliers, but most are much higher (2200uF & 220uF) than TI recommendation in the PGA2311 datasheet (10uF & 0.1uF). What are the purpose of these? To supply instant current drain for audio peaks?
The audio input and output resistors In some designs these are accompanied by a X uF cap on each channel (to reduce input DC offset?). There are 220R input resistors and 330R output resistors. 220R at the input might be to raise input Z, but why would you want to increase output Z? And there are two 10k to ground - Why would you want to short circuit to ground?
Last, the Digital power supply of the PGA(?). This seems to be taken from the analog +5V, via 10R to pin 1 and 4 and on to ground via 10uF cap. In other designs this is taken from VCC and not analog +5V. What does the 10R and the 10 uF cap here? Are there any practical effects of supplying the digital with the "theoretical" analog power, than the "theoretical" digital power?
By "other designs" I'm mostly referring to this alternative schematic of the same chinese preamp: http://www.platenspeler.com/diy/preamps/remote/pga2311_schema.jpg
View attachment 323144 [/QUOTE]
The power supply caps are often sized as 'this sounds about right' or 'this is what I had handy'. Sounds funny but its not as bad as it seems. The PGA draws so little power that anything above 1000uF or so gets you in diminishing returns.
Instant current drain? A signal of 5V into 10k requires 0.5mA - hardly a big current drain. The 10uF/0.1uF in the data sheet are for local decoupling, and assume that the power supply itself has been taken care of.
The 220ohms on the output side are required to protect the chip when the mute relays are engaged - they short the output to gnd. The 330 input are a (limited) protection of the input to external junk like people touching the signal input after having been sitting on nylon chairs...
A smallish (few 100) ohms on the output is a good idea to protect the preamp from accidental shorts and from too large cap loads (which might cuase instability). As a preamp designer, you have no control over whatever anybody hangs off of it, so you need to play better safe than sorry.
The 10 ohm to 10uF provides some additional smoothing for the supply to the digital supply pin. This is not necessary for the digital pin but might have been put in to avoid backtalk of digital hash into the main supply. Then again, it might have been put in because 'everybody does it' so it does look good. Don't look so surprised!
jan
Last edited:
Thanks Jan for your very educational response, that's about what I had imagined. But I'm still wondering what the power input decoupling(right term?) caps are doing. What would happen if I removed them? What are their purpose in life, living in a little preamp box
Shouldn't the designer have added the local decoupling caps, as suggested by TI? Or is it already taken care of by the "main" caps? Might perhaps be if integrating the PGA in a larger circuit, such as an integrated amp.
Shouldn't the designer have added the local decoupling caps, as suggested by TI? Or is it already taken care of by the "main" caps? Might perhaps be if integrating the PGA in a larger circuit, such as an integrated amp.
Thanks Jan for your very educational response, that's about what I had imagined. But I'm still wondering what the power input decoupling(right term?) caps are doing. What would happen if I removed them? What are their purpose in life, living in a little preamp box
Shouldn't the designer have added the local decoupling caps, as suggested by TI? Or is it already taken care of by the "main" caps? Might perhaps be if integrating the PGA in a larger circuit, such as an integrated amp.
Well you need some fairly large cap after the rectifier to convert the half-sinewaves from the rectifiers to DC with some ripple.
Then the regulators finish the job and actually you need caps after the regulators for regulator stability, although 10uF normally is enough. The reg output Z is so low that the cap after the reg doesn't really do a lot to quiet the supply (except the stability as mentioned).
Well spotted that local decoupling. It is always good to that as close as you can get to the chip, as recommended in the data sheet. Are you sure it isn't there? Sometimes the schematic doesn't show all the parts on the actual board.
The local decoupling with a good cap is better than an expensive film cap at the regulator -the wiring inductance may make that expensive cap all but invisible from the chip.
jan
It's not there. Should be quite easy to add though. One or two caps parallelled from the AV+ and AV- pins to any ground space on the pcb i can find? Or need they connect right at the PGA ground pin?
Edit: Forgot to ask: The two 10K resistors to ground, blue marked, what do they do?
Edit: Forgot to ask: The two 10K resistors to ground, blue marked, what do they do?
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.