I'm making a DIY guitar widget...my head is polluted with all manner of jumble & I'm now hitting the 'wall' & need someone to extend their hand out & help me over it (then I'll be off again!!).
Here's the signal chain...
1st stage - Non Inverting opamp (Voltage gain of 10) ->
2nd stage - AGC controlled non inverting opamp (voltage gain between 2 & 4)
3rd stage - low power audio amp chip (TDA7052A ...I can use anything here, it only needs about 300mW+ of power)
4th Stage - an 8 ohm driver coil (to stimulate the strings - don't ask!!!)
This all presently works off a 5V single supply rail (therefore the AC guitar signal has a theoretical 2.5V headroom either side of the virtual earth which is sitting at 2.5V). I'm using rail to rail opamps.
Ok then, so with an input of 100mV peak to peak, the guitar signal exits the first opamp stage at about 1V peak to peak. It feeds into the next stage, where it exits at between 2V & 4V (depending on what level I set the AGC to be)
Here's my problem (I have others, but I'll just put this one forward for now!)...
1. when the TDA7052A power amp stage sees an input that's too big (circa 2.2V, it's output goes all skewiff (almost as if the negative portion of the signal have been flipped up - fullwave rectifier-ish). weirdly it's not clipping (which is what I'd have expected if I was feeding the amp too large a signal).
This begs the question - with these 'all in one' monolithic audio amp chips - am I constained to a maximum input voltage (I've checked the TDA7052A datasheet & saw no such data). Are their 'known' quirks with such chips if feeding them too big an input (aside from 'clipping')
The TDA7052A chip does add its own gain (which is adjustable - but even at it's lowest setting, it'll always add a default amount of gain) & I guess if my input is too large it will amplify the input & possibly hit 'problems' - would I be right in assuming that I should derive the maximum input voltage to the power amp chip by the following....
Maximum input = Rail voltage / power amp's default gain.
in other words, if the amp applies a gain of say 2 & the rail is 5V, then I should not feed it an input any larger than 2.4V peak to peak? would that be right?
in order to get enough 'granularity' from the 'level detect' aspect of my AGC circuit, I really would have liked to keep that 2nd stage output (ie the input to the power amp) at about 4V peak to peak....so what are my options wrt the power amp stage here?
ie what would be a good solution for my power amp stage - perhaps something like unity gain voltage to current device? (which I guess would rule out a nice simple tidy monolithic amp chip?)
My only constarint is that the output stage should now be a current hog (therefore no class A stuff!)
Many thanks for your help in anticipation
Here's the signal chain...
1st stage - Non Inverting opamp (Voltage gain of 10) ->
2nd stage - AGC controlled non inverting opamp (voltage gain between 2 & 4)
3rd stage - low power audio amp chip (TDA7052A ...I can use anything here, it only needs about 300mW+ of power)
4th Stage - an 8 ohm driver coil (to stimulate the strings - don't ask!!!)
This all presently works off a 5V single supply rail (therefore the AC guitar signal has a theoretical 2.5V headroom either side of the virtual earth which is sitting at 2.5V). I'm using rail to rail opamps.
Ok then, so with an input of 100mV peak to peak, the guitar signal exits the first opamp stage at about 1V peak to peak. It feeds into the next stage, where it exits at between 2V & 4V (depending on what level I set the AGC to be)
Here's my problem (I have others, but I'll just put this one forward for now!)...
1. when the TDA7052A power amp stage sees an input that's too big (circa 2.2V, it's output goes all skewiff (almost as if the negative portion of the signal have been flipped up - fullwave rectifier-ish). weirdly it's not clipping (which is what I'd have expected if I was feeding the amp too large a signal).
This begs the question - with these 'all in one' monolithic audio amp chips - am I constained to a maximum input voltage (I've checked the TDA7052A datasheet & saw no such data). Are their 'known' quirks with such chips if feeding them too big an input (aside from 'clipping')
The TDA7052A chip does add its own gain (which is adjustable - but even at it's lowest setting, it'll always add a default amount of gain) & I guess if my input is too large it will amplify the input & possibly hit 'problems' - would I be right in assuming that I should derive the maximum input voltage to the power amp chip by the following....
Maximum input = Rail voltage / power amp's default gain.
in other words, if the amp applies a gain of say 2 & the rail is 5V, then I should not feed it an input any larger than 2.4V peak to peak? would that be right?
in order to get enough 'granularity' from the 'level detect' aspect of my AGC circuit, I really would have liked to keep that 2nd stage output (ie the input to the power amp) at about 4V peak to peak....so what are my options wrt the power amp stage here?
ie what would be a good solution for my power amp stage - perhaps something like unity gain voltage to current device? (which I guess would rule out a nice simple tidy monolithic amp chip?)
My only constarint is that the output stage should now be a current hog (therefore no class A stuff!)
Many thanks for your help in anticipation
Ok, at the risk of talking to myself(!), I've just realised the way out of this...& it's stupendously, embarrasingly simple 
....
put a preset to ground from the second opamp output stage, taking the preset wiper into the poweramp.
this way I can have the full 4V'swing' to keep my AGC running sweet, but I can also tame the signal going into the poweramp chip.
How does that happen...ie the moment you ask the question...you suss the answer immediately afterwards (just a shame it's such a terribly simple solution - like I say, my head is full of jumble & I can't see the wood for the trees!)
[puts coat on, to go & hide in the corner]
....put a preset to ground from the second opamp output stage, taking the preset wiper into the poweramp.
this way I can have the full 4V'swing' to keep my AGC running sweet, but I can also tame the signal going into the poweramp chip.
How does that happen...ie the moment you ask the question...you suss the answer immediately afterwards (just a shame it's such a terribly simple solution - like I say, my head is full of jumble & I can't see the wood for the trees!)
[puts coat on, to go & hide in the corner]
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.