Slightly odd question;
When using the PGA series digital volume control IC's is there any noticeable delay in varying the volume, any audio artifacts? or is the behavior much like a conventional resistive control?
My application is that I want to use them in a DJ idolator (type of EQ) but need the response to be percived as instant and smooth. The digipots are attractive as they avoid running the audio signal through the pot and reduce the requirments of the control pots.
When using the PGA series digital volume control IC's is there any noticeable delay in varying the volume, any audio artifacts? or is the behavior much like a conventional resistive control?
My application is that I want to use them in a DJ idolator (type of EQ) but need the response to be percived as instant and smooth. The digipots are attractive as they avoid running the audio signal through the pot and reduce the requirments of the control pots.
PGA series chips are intended for audio. There is zero cross detection so no ticks
when changing level. The steps are 1/2 dB so with the 8 bit control the spread is
128 dB ranging from -95.5 up to +31.5. They will react as fast as you can send
data to them and they will clock up to 6.25 MHz for loading data. Do you mean
'digipots' like Analog Devices? These are very good but I don't know if they have
log taper and the zero cross detecion thing to tailor it for audio use. I have used
the PGA2311 (same as Crystal Semiconductor CS3310) in my TV audio both the
stereo and 5.1 systems. There are never any anomolies. I run them at 2 steps per
change meaning 1 dB steps and have digital readout out the PGA drive showing
1 dB steps. Channels track less than 0.1 dB error IE you will not hear it.
Being .5 dB per step you can easily do channel balance by offsetting data to the
individual channels so a channel can be N dB different from the others and it will
track.
I use a lowly Arduino to load them and just simply 'bit bang' the data and clock so it
is really simple to do. Arduino listens to a shaft encoder more local control, IR remote
and loads the readout display. Overall an easy project from a programming standpoint.
Happy to give you Arduino code to see one way to do it
G²
Thanks for your reply stratus, yes I am talking abou the PGA series chips from TI, example code would be great. I haven't decided yet whether to use the +/-15V or +/-5V chips, some input processing is required to accept balanced +24dBu signals and consumer level signals regardless so the choice is more about the performance of the overall device when operating at the different internal signal levels. From the datasheet of the PGA2311:
The SPI interface delay is negligible operating at MHz frequencies. so even with a worst case of 16mS delay response to POT changes they should be perceived as instant? the use case is this:
The SPI interface delay is negligible operating at MHz frequencies. so even with a worst case of 16mS delay response to POT changes they should be perceived as instant? the use case is this:
The data is clocked into the register and as soon as chip select goes high the
gain goes to the new setting. I run the TVs mute as automatic ramp up/
ramp down so new values keep getting updated, I normally use 50 dB/second
ramp rate and the audio smoothly changes level.
If you're worried about levels the high voltage version is just the same functionally,
just a lot more headroom. I worked in the broadcast world and while balanced anlog
audio was the norm for interconnects, the audio was always converted back to single
ended internally BUT if you want to run parallel PGA chips for balanced throughout it
would work just fine, only cost more for parts. Once the signals have gone digital,
conversoin back to analog will not have any peaks above FS regardless so using the
2311 5V chip wll never hit an input overload but at max gain certainly could clip.
The 2310 would give about 10dB extra output.
I've included the latest version of the software for the TV remote. The Setgain(),
trimvol () and lod8vol() bit-bang loads the 3 PGA2311 chips to do all 6 channels.
There are similar routines to load the NJW1119 tone control chips. Readknob()
looks at the shaft encoder for local volume/mute change. Slowmute() actually
ramps the levels up/down. It 'listens' to Samsung and LG remotes.
The first 200 lines are comments describing the changes to the next version.
The .txt of the file gets changed to .ino for the arduino. DIY used to flag .ino files
as bad stuff when uploading files.
G²