This is my first post on these forums, but as I am just finishing uni and am very interested in the field of audio electronics I suspect It will be far from my last, so Hello.
I'm currently designing a valve pre-amp for guitar with built embedded system featuring in FPGA driven DSP offering a range of effects, using the TLV320AIC23B codec. I'm currently considering the DAC output of the codec (which has 1Vrms range), the signal from which will be the line-out signal of the system. I feel as though I should add some kind of buffer to this output , rather than tie the expensive codec DAC output directly to the system line out.
Given the fact that I'm a bit of a newbie with this kinda stuff, I thought I would post on here to see if anyone has any recommendations for a single rail OPAMP which I can use as the line-out buffer.
Ideally the OPAMP will be rail-to-rail, able to run off VCC as low as 3v and have some kind of built-in short-circuit condition protection. I will of course be looking into it myself, but any suggestions are welcome.
Cheers, Chris.
I'm currently designing a valve pre-amp for guitar with built embedded system featuring in FPGA driven DSP offering a range of effects, using the TLV320AIC23B codec. I'm currently considering the DAC output of the codec (which has 1Vrms range), the signal from which will be the line-out signal of the system. I feel as though I should add some kind of buffer to this output , rather than tie the expensive codec DAC output directly to the system line out.
Given the fact that I'm a bit of a newbie with this kinda stuff, I thought I would post on here to see if anyone has any recommendations for a single rail OPAMP which I can use as the line-out buffer.
Ideally the OPAMP will be rail-to-rail, able to run off VCC as low as 3v and have some kind of built-in short-circuit condition protection. I will of course be looking into it myself, but any suggestions are welcome.
Cheers, Chris.
Last edited:
1 vrms and 3 volt supply is a tough requirement.
I would suggest you look at the TI (Texas Instruments) and Linear Tech sites and use the parametric search functions to see if anything fits your requirements. You might find you need to work with surface mount parts to meet your specs.
I would suggest you look at the TI (Texas Instruments) and Linear Tech sites and use the parametric search functions to see if anything fits your requirements. You might find you need to work with surface mount parts to meet your specs.
Candidates for scrutiny include the National LM2900, 3900, 3901, On Semi MC33201 33202, 33204, Linear Technoloty LT1211,LT1212
The National parts were used in a lot of organs and tend to have significant failure rate about the 10 year point.
On semi parts are slow
There are various limits of each one, read datasheets available from datasheetcatalog.com
After looking through data and finding no stock on On Semi parts I bought two zener diodes and two resistors and made a dual supply out of a single. Also requires capacitor coupling in and out of the op amp and a biasing pair of resistors on the input.
The National parts were used in a lot of organs and tend to have significant failure rate about the 10 year point.
On semi parts are slow
There are various limits of each one, read datasheets available from datasheetcatalog.com
After looking through data and finding no stock on On Semi parts I bought two zener diodes and two resistors and made a dual supply out of a single. Also requires capacitor coupling in and out of the op amp and a biasing pair of resistors on the input.
In the Digital World *everything* runs on 5V or 3.3V single supply.
Fine ... while you stay there.
In the Analog World in principle everything runs on +/-15V , so consider the extra winding (s) and simple inexpensive supply components needed to get them.
To boot, as Mooly noticed, no way you can get even1V RMS out of a single 3.3V supply and even if you barely did, it´s not clean unadultered signal but "clipping* signal, since you will have zero headroom.
Going up to 5V will give you, what? 1 or 2 dB headroom?
In practice same as none at all.
So in a nutshell: do your digital magic in the 3 or 5V world, but once you get analogic, supply it what it needs.
Not a big deal anyway and to boot .... you already got *hundreds* of volts for the Tubes, didn´t you? 😉
Last edited:
Texas TLV2374 are rail to rail, 2.7 to 16V and 3MHz GBW and low supply current.
They also drive 600R nicely, something most rail to rail don't
They also drive 600R nicely, something most rail to rail don't
TLV2374 has terrible THD+N and noise numbers. I would suggest to examine the power supply input. If it is somewhere around 5VDC and then regulated down to 3.3V for the FPGA then it would be possible to tap off from 5VDC and regulate down to 4.5VDC (using one of the new LDO regs) which is enough for many new op-amps.
OPA2322 is a good option, or OPA2316 (both dual op amps).
Whichever op amp you chose, I'd use an inverting configuration. You can bias the input common mode voltage using the non inverting pin. And the inverting configuration avoids input crossover distortion which is common in classic rail to rail input stages.
Whichever op amp you chose, I'd use an inverting configuration. You can bias the input common mode voltage using the non inverting pin. And the inverting configuration avoids input crossover distortion which is common in classic rail to rail input stages.
I use the Microchip MCP6022 and/or the Linear Technology LTC6246 when I need true rail-to-rail operation with (+3.3, GND) supplies.
Wow, thanks for all the suggestions guys, really appreciate all of them.
My apologies for the slow reply, so busy at the moment, its non-stop for the next few weeks! I'm currently working on my FPGA design, valve pre-amp, and power supply for the whole lot, but in a few days when I'm happy with those bits I'll turn my attention back to the codec buffers and update!
Cheers, Chris.
My apologies for the slow reply, so busy at the moment, its non-stop for the next few weeks! I'm currently working on my FPGA design, valve pre-amp, and power supply for the whole lot, but in a few days when I'm happy with those bits I'll turn my attention back to the codec buffers and update!
Cheers, Chris.
- Status
- Not open for further replies.