It depends on several factors, and I had also mentioned that.
1) DC offset of the opamps you use. With OPA627 it shows 1.6mV on LTS. With OPA1656 it goes up 20mV. The latter might need opening the bypass. A turn on thump from the pure-DC preamp is certain every time, which could be muted with a switch, or simply turning the preamp before the power amp or without the preamp selected. Or you use a delay turn-switch on the preamp or the power. We already did have that care in the old preamp/power amp days. More expensive units already provided that delay.
2) Checking if you have an input cap on the next stage. If you do have a cap it should stop any DC offset. Audio designer Ben Duncan said something interesting once. As long as the offset does not exceed 1V on each stage, you should use just one only cap at the power amp stage, and that should be a top quality film type, polypropylene or better.
In my case I do no plan to listen to LP discs on my receiver, at least at first. My ideia is to capture all my LP collection using a professional Tascam record, which has excellent preamps, and then to clean all surface noise and pops digitally on the PC.
This would be the second time I do that, though my first time was quite promising. I used a Behringer UCA-202 small line preamp, that worked quite well. The RIAA preamp used to feed the Behringer was a Luxman discrete preamp.
The Behringer only worked at 16-bit/48kHz, which gave me a smaller file to correct the noise on, and this time I plan to record larger files, with more resolution.
For digital processing I will be using Audacity again. It's a very powerful and easy to use professional audio program, and it will do all I need.
1) DC offset of the opamps you use. With OPA627 it shows 1.6mV on LTS. With OPA1656 it goes up 20mV. The latter might need opening the bypass. A turn on thump from the pure-DC preamp is certain every time, which could be muted with a switch, or simply turning the preamp before the power amp or without the preamp selected. Or you use a delay turn-switch on the preamp or the power. We already did have that care in the old preamp/power amp days. More expensive units already provided that delay.
2) Checking if you have an input cap on the next stage. If you do have a cap it should stop any DC offset. Audio designer Ben Duncan said something interesting once. As long as the offset does not exceed 1V on each stage, you should use just one only cap at the power amp stage, and that should be a top quality film type, polypropylene or better.
In my case I do no plan to listen to LP discs on my receiver, at least at first. My ideia is to capture all my LP collection using a professional Tascam record, which has excellent preamps, and then to clean all surface noise and pops digitally on the PC.
This would be the second time I do that, though my first time was quite promising. I used a Behringer UCA-202 small line preamp, that worked quite well. The RIAA preamp used to feed the Behringer was a Luxman discrete preamp.
The Behringer only worked at 16-bit/48kHz, which gave me a smaller file to correct the noise on, and this time I plan to record larger files, with more resolution.
For digital processing I will be using Audacity again. It's a very powerful and easy to use professional audio program, and it will do all I need.
a) is better than b), noise-wise. The circuit you rejected is better than both, noise-wise.
Assuming a good FET op-amp, the first two circuits of post #457 increase the equivalent input noise current from 0.5869 pA/√Hz (noise current of a 47 kohm termination resistor) to 1.03 pA/√Hz. That's a 4.885 dB increase of the main noise contribution. It will definitely be audible in an A-B comparison with no record playing, it may or may not be audible as a somewhat sharper sounding background noise when there is a very clean record playing.
No, it doesn't, actually. The feedback makes the voltage at the negative op-amp input an accurate copy of the voltage at the positive input, no matter what value R3 has and no matter whether there is a capacitor in series with it, so R2 sees a very low impedance on its right side.
There is an imperfection, though: the current that flows through R2 into the feedback network of the op-amp shifts a zero that should be in the origin slightly into the right half plane. Its effect is negligible as long as R2 is much greater than R3, which it needs to be anyway.
Looks like a good solution to me, as long as the subsonic rumble is far below the level that would drive the first stage into clipping.
replace U2 OPA627/1656 with OPA192 and the problem will disappear. I recommend replacing U1 with OPA192 as well - it will be 10 times cheaper and 5 times better.
Last edited:
I already have some OPA192 samples that I will listen to in the three preamps versions I will build for audition comparison. Several opamps will be compared.
I use OPA192 since 2022 in all my preamp versions 5...9 - a nice SOIC-8 with 4 nV/sqrtHz and <5 uV DC offset
Attachments
I found exactly the same SMD preamp you are using being sold through Facebook, and it seems it uses the OPA192 as a DC servo, not for the whole RIAA preamp. It's a dual JFET input discrete preamp, very good design BTW.
But I have checked again the OPA192 on the RIAA three versions I simulated, and offset specs seem very low, like a mere 115uV.
But I have checked again the OPA192 on the RIAA three versions I simulated, and offset specs seem very low, like a mere 115uV.
Yes I use OPA192 as DC servo, but your output offset 115 uV (or approx. 5 uv x 25 DC gain = 125 uV) is also very good for "no electrolytics" circuit design
Which is the best/cheaper place to get OPA1656 converted to DIP8?
I am not sure those sold by Aliexpress are not fake.
I am not sure those sold by Aliexpress are not fake.
From #287
With all that said and in my small world, I can't imagine ANY phono preamp devoid of "sharp noise" if it's got a 47k cart load. With the 47k contributing 0.5869 pA/√Hz it then rises to a scary 1.03 pA/√Hz with the inclusion of R2's noise current. Now we really have a whole lot of "sharp noise" that nobody's noticed before. Who's heard of "sharp noise"? Is it ever addressed in the ads, brochures and reviews of the finest and biggest $$$ phono preamps? Do consumers of unknown numbers of MM preamps hear it..? You say it is perceivable. It's just there when no LP is playing and is annoying. Hmmmm.....
There certainly are people who get annoyed by the 3.4 pA/√Hz of an LT1028. I've seen a thread on this forum from someone who had made a MM phono amplifier with an LT1028 because he did not know about current noise, and then got very disappointed when he heard the hiss. Besides, @kgrlee has also written about the poor results of LT1028-based amplifiers in double-blind tests. (Then again, there are also people who built the Elektor Supra 2.0 MM phono preamplifier with four LT1028s and are happy with it, despite its 6.8 pA/√Hz equivalent input noise current. The moving-magnet Supra 2.0 is the worst design I have ever seen.)
Of course 1.03 pA/√Hz is some 10 dB better than 3.4 pA/√Hz. Whether that is bad enough to annoy people during normal use, I don't know. I do know it is about 4.8 dB worse than a conventional design with a FET op-amp, or than my single-loop design.
Regarding an A/B test without record playing: assuming that cartridge thermal noise (including iron losses) and the other noise sources together contribute about as much noise as the 47 kohm, you will have a 3 dB noise floor increase when R2's noise is added. There are people who can hear 0.2 dB changes under double-blind conditions, so I would be surprised if 3 dB went unnoticed.
The noise current gets converted to voltage across the impedance of the (loaded) cartridge and it then gets filtered by the RIAA correction. Due to the large increase of the cartridge impedance with frequency, you end up with noise that has most of its power at high frequencies, which I described as sharp noise.
Of course 1.03 pA/√Hz is some 10 dB better than 3.4 pA/√Hz. Whether that is bad enough to annoy people during normal use, I don't know. I do know it is about 4.8 dB worse than a conventional design with a FET op-amp, or than my single-loop design.
Regarding an A/B test without record playing: assuming that cartridge thermal noise (including iron losses) and the other noise sources together contribute about as much noise as the 47 kohm, you will have a 3 dB noise floor increase when R2's noise is added. There are people who can hear 0.2 dB changes under double-blind conditions, so I would be surprised if 3 dB went unnoticed.
The noise current gets converted to voltage across the impedance of the (loaded) cartridge and it then gets filtered by the RIAA correction. Due to the large increase of the cartridge impedance with frequency, you end up with noise that has most of its power at high frequencies, which I described as sharp noise.
I have prepared a real world record of the noise, maybe for @JRA (even if I doubt he is really interested in the roots of the issue), maybe for someone else who might be interested.
A simple RIAA preamp with one opamp and 40dB/1kHz gain has Shure M35X cartridge at the input. The output noise is recorded with 2 opamps, LT1028 and OPA 627. Bellow in the zip file are 2 files with recorded noise. 48kHz sampling, 32-bit float wav mono files. The FS level corresponds to 500mVrms. So, give it a chance, listen and you will know what is a "sharp" noise.
P.S.: adding file statistics
Last edited:
I think by now (more than three decades later) everyone has come to the realization that the LT1028 is not the first choice as an MM-RIAA input'thing - and anyone who read and understood the data sheet in printed paper form back then already knew that.
It all depends on how you use this fantastic component. This strange magazine called Elektor ran the Belcanto preamplifier article...
It all depends on how you use this fantastic component. This strange magazine called Elektor ran the Belcanto preamplifier article...
There is nothing wrong with the files. You only need a proper tools to play them. Foobar, Audacity - both work, speaking about free tools. You must be able to play 32 bit mono files. Cell phones probably would not work. First check your system (and settings) before saying that there is something wrong with the files. And, obviously, please turn volume up. It is the background noise, not music.
Anyone else cannot play them?
Anyway, I have just converted them to 24-bit and attached below. I will NOT make 16-bit files of them, because they would be degraded by both quantization noise and dithering.
Last edited:
Sorry, no audio is heard.
And I had used Foobar, Audacity and Sound Forge to play them, before and now.
But... there seems to be something recorded on the track the Sound Forge shows on the meter, only it's down -60dB
And I had used Foobar, Audacity and Sound Forge to play them, before and now.
But... there seems to be something recorded on the track the Sound Forge shows on the meter, only it's down -60dB
Just increased the volume in Sound Forge, 20dB, two times, and there's some noise there indeed.