Hi all,
I'm looking to start a project for a portable phono preamp that can be USB charged, which means I'll be dealing with battery voltages as low as 3V. From what I've gathered, this will be tough to design in a way that has much headroom/overload. Is this tilting at windmills, or possible if I use active RIAA eq and carefully design my stages? There's also the problem that this is a single ended source, would I want to bias the MM cart or would I be better off making a negative rail (that somehow isn't noisy)?
I'm new to this, although I earned my B.Sc. in Electrical Engineering in 2018. I think I can handle the RIAA transfer function, but from a few hours of browsing, it's apparent that there is a lot more going on. My general idea is to use OPA1692's for the preamp and make life easy with a TPA6138a2 for headphones.
Am I on the right track? Did I get it right the first time? (ha)
If you could point me to the required reading, or offer any help at all, I'd be grateful
I'm looking to start a project for a portable phono preamp that can be USB charged, which means I'll be dealing with battery voltages as low as 3V. From what I've gathered, this will be tough to design in a way that has much headroom/overload. Is this tilting at windmills, or possible if I use active RIAA eq and carefully design my stages? There's also the problem that this is a single ended source, would I want to bias the MM cart or would I be better off making a negative rail (that somehow isn't noisy)?
I'm new to this, although I earned my B.Sc. in Electrical Engineering in 2018. I think I can handle the RIAA transfer function, but from a few hours of browsing, it's apparent that there is a lot more going on. My general idea is to use OPA1692's for the preamp and make life easy with a TPA6138a2 for headphones.
Am I on the right track? Did I get it right the first time? (ha)
If you could point me to the required reading, or offer any help at all, I'd be grateful
You can use a DC-DC converter after the battery and go with a typical dual rail supply everywhere. Nowadays battery capacitance is not a big problem so you can sacrifice a bit of power efficiency and design a better performing circuit. Also the choice of op-amps and headphone amp ICs will be far greater if you use standard supply voltages.
Regards,
Oleg
Regards,
Oleg
> battery voltages as low as 3V. ... use OPA1692
Suggested for 3.5V up. Are you confident about operation at your intended voltage?
The usual RIAA in the NFB has trouble because RIAA rolls-off forever, and the NFB configuration tends to gain of unity. Working at gain of 50, the error happens as we approach 100KHz, so not an issue (except for lily-guilders). Working at much lower gain the error is appreciable.
Since speech/music power falls in the top octaves, we can defer the top pole. Since the '1692 can pull a load, we can use a passive post-filter to get the correct response.
The plan below has midband gain of 11. Since a gain of 50-100 works OK (ample headroom) with 30V amplifiers, this should be OK with a 3V amplifier. Nominal output like 55mV (with peaks of 400mV, 1.2V peak-peak). This can be followed with a volume pot, then a gain of 10 to give headphone levels suitable for a 3V supply.
Suggested for 3.5V up. Are you confident about operation at your intended voltage?
The usual RIAA in the NFB has trouble because RIAA rolls-off forever, and the NFB configuration tends to gain of unity. Working at gain of 50, the error happens as we approach 100KHz, so not an issue (except for lily-guilders). Working at much lower gain the error is appreciable.
Since speech/music power falls in the top octaves, we can defer the top pole. Since the '1692 can pull a load, we can use a passive post-filter to get the correct response.
The plan below has midband gain of 11. Since a gain of 50-100 works OK (ample headroom) with 30V amplifiers, this should be OK with a 3V amplifier. Nominal output like 55mV (with peaks of 400mV, 1.2V peak-peak). This can be followed with a volume pot, then a gain of 10 to give headphone levels suitable for a 3V supply.
Attachments
The problem of noise from the DC-DC converter is well worth avoiding. Plenty of rail-rail opamps will run from 3.7V quite happily, I'd suggest AD8656 since it is low current noise being CMOS and can drive loads of current for headphones.
I would not be worried much about the noise of DC-DC converters. If done right DC-DC converter based PSU can be very quiet. Look at the Silent Switcher by Jan Didden. I've also designed and built a DC-DC module based PSU and then a DC-DC converter powered (using USB charger) headphone amp. The measurements show no visible effects of the converter noise on the audio output.
As for the AD8656 suggestion, I see a potential problem for the headphone duty. If I understand Fig. 10 in AD8656 datasheet correctly high output current results in significant reduction of the already low supply overhead voltage. In practical terms using 3.7 VDC single rail supply the output will be limited to about 1Vrms at 50mA output current and will clip badly at higher levels. This will be especially pronounced on bass notes as is typically observed in power amps.
Regards,
Oleg
As for the AD8656 suggestion, I see a potential problem for the headphone duty. If I understand Fig. 10 in AD8656 datasheet correctly high output current results in significant reduction of the already low supply overhead voltage. In practical terms using 3.7 VDC single rail supply the output will be limited to about 1Vrms at 50mA output current and will clip badly at higher levels. This will be especially pronounced on bass notes as is typically observed in power amps.
Regards,
Oleg
Thanks for the help so far everyone,
I started the schematic layout today. I've used PRR's NFB schematic, swapping in the OPA1942 since it has JFET inputs, which should be better for the high-impedance input. I'm a little iffy about linking it to the TPA, since the TI app note is a differential filtering amplifier. I'll be ok if I just stick a large film capacitor between the PRR amp and the non-inverting input of the TPA and setting gain with the feedback, right? I feel like this is a very simplistic design, but I guess I'm trying to solve a simple problem. I'd really like this amp to be crystal clear with good bass response, so I imagine using these high-quality components and big caps should get me close.
Next steps are looking into cartridge loading and the power supply. I'm thinking split rail +/- 5V would efficiently convert from lithium or USB and the TPA can run off a linear reg.
I started the schematic layout today. I've used PRR's NFB schematic, swapping in the OPA1942 since it has JFET inputs, which should be better for the high-impedance input. I'm a little iffy about linking it to the TPA, since the TI app note is a differential filtering amplifier. I'll be ok if I just stick a large film capacitor between the PRR amp and the non-inverting input of the TPA and setting gain with the feedback, right? I feel like this is a very simplistic design, but I guess I'm trying to solve a simple problem. I'd really like this amp to be crystal clear with good bass response, so I imagine using these high-quality components and big caps should get me close.
Next steps are looking into cartridge loading and the power supply. I'm thinking split rail +/- 5V would efficiently convert from lithium or USB and the TPA can run off a linear reg.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.