HQ PU RIAA PREAMP
Revive your vinyls
You might have an old Dual, Lenco, Thorens turntable, ... now is the time to bring it back to life with a guarantee of quality.
The diagram shown is taken from an old AN-124 application note from PMI Analog Devices around SSM2015. The latter being obsolete, I used a THAT1510.
The peculiarity of this assembly lies first in a pre-amplification before attacking the RIAA filter. Usually, a lot of PAs are built around a single operational amp (NE5534). As the noise level of the That 1510 is significantly lower, you will notice the difference. It will always be possible to modify the gain with the formula G = 1 + 10K / Rg
Each cell has its attributes and the level is sometimes very different. If one wants at That 1510, the gain resistance can be replaced by a fixed 100 ohm resistor in series with an adjustable 220 ohm. The gain will be variable from 30 to 40 dB.
The NE5532 of the filter and the output stage can be replaced by LM4562
Discussion (11 comments)
Marc Schwarzburg 3 years ago
It is not so hard to do the correct Riaa deemphasis. Lipshitz and Jung show us a decent solution in an old AES Journal from the 1980ties.
Its totally passive like the ones you find in the tube phono preamps from the 1960 but with low impedance since we use semiconductors here. You can expect lower phase shifts in the filter than in solutions within the Op Amp feedback loop as other companies did...
You should use 0.1 % metalfoil resistors and 1% mkp capacitors in the filter or you select them from the junkbox after measuring some parts to get the right values.
the network will be placed between the output of "that1510" Opamp and the following input of the output buffer stage.
For the that1510 gain resistor i would recommend an 22 Ohm and a switch for paralleling an 220 Ohms resistor.
So you can choose between MM or MC systems.
In every case you will need a super clean power supply with no ripple on the rails. (shunt regulator or simple dual battery setup)
Its totally passive like the ones you find in the tube phono preamps from the 1960 but with low impedance since we use semiconductors here. You can expect lower phase shifts in the filter than in solutions within the Op Amp feedback loop as other companies did...
You should use 0.1 % metalfoil resistors and 1% mkp capacitors in the filter or you select them from the junkbox after measuring some parts to get the right values.
the network will be placed between the output of "that1510" Opamp and the following input of the output buffer stage.
For the that1510 gain resistor i would recommend an 22 Ohm and a switch for paralleling an 220 Ohms resistor.
So you can choose between MM or MC systems.
In every case you will need a super clean power supply with no ripple on the rails. (shunt regulator or simple dual battery setup)
Reply
ERRYSON 3 years ago
Hello,
I see that this is a bit the principle adopted in the manual of LM4562 or LME49720.
I see that this is a bit the principle adopted in the manual of LM4562 or LME49720.
Reply
Marc Schwarzburg 3 years ago
thank you Erryson for pointing that out !
Yes this is a proven solution for high end audiophile equipment. Some famous phono preamps are based on this principle. I´ve seen this in even 4K USD equipment because of the linearity of the implementation.
The deviation in the parts of LME49720 datasheet is to compensate the parameters of this particular opamp shown.
btw. this Opamp should be awesome. I should really check this device.
My 2nd opamp is an Burr Brown OPA134 and this is excellent too. Don´t forget to block the supply pins with 220nf foil and 470uF electrolyt caps in parallel - nearest to the chip.
I attached a example schematic I found last year somewhere , I forgot where... but had a hardcopy
Yes this is a proven solution for high end audiophile equipment. Some famous phono preamps are based on this principle. I´ve seen this in even 4K USD equipment because of the linearity of the implementation.
The deviation in the parts of LME49720 datasheet is to compensate the parameters of this particular opamp shown.
btw. this Opamp should be awesome. I should really check this device.
My 2nd opamp is an Burr Brown OPA134 and this is excellent too. Don´t forget to block the supply pins with 220nf foil and 470uF electrolyt caps in parallel - nearest to the chip.
I attached a example schematic I found last year somewhere , I forgot where... but had a hardcopy
Reply
ERRYSON 3 years ago
Hello,
It is possible to reduce the low end of the audio spectrum by inserting a capacitor with RG That1510 (2200µF 6V3 Muse Nichicon or Elna style)
For the OPA2134, it is a good output circuit because I have seen in the past some diagrams associating with output transformers JT112 of Jensen.
It is possible to reduce the low end of the audio spectrum by inserting a capacitor with RG That1510 (2200µF 6V3 Muse Nichicon or Elna style)
For the OPA2134, it is a good output circuit because I have seen in the past some diagrams associating with output transformers JT112 of Jensen.
Reply
Marc Schwarzburg 3 years ago
usually you do the low end after the Riaa network or after the 2nd opamp buffer as I see at a german design.
Output transformer with opamps ? why ? whats the reason for this ? you will lost bandwith with bass and treble (high tones) because of eddy current in the iron core. I think that is stupid except you will need a studio balanced / symetrical XLR output. btw a good Audiotransformer is expensive.
Output transformer with opamps ? why ? whats the reason for this ? you will lost bandwith with bass and treble (high tones) because of eddy current in the iron core. I think that is stupid except you will need a studio balanced / symetrical XLR output. btw a good Audiotransformer is expensive.
Reply
Show more
1 Attachment(s)
1 Comment(s)
Club AC-News 3 years ago
The reviews are disappointing. The author has at least created something as a basis for work. This base is perhaps the beginning of a future evolution. Who asked the question? An achievement, in itself, is never finished and always evolves over time. If that's not right for you, who's ready to step up and step up?
Reply
ERRYSON 3 years ago
Thank you AC NEWS,
You understand my approach because this U-shaped sheet is used to bring together the connections, the 2 L&R preamps, the buffered stereo RIAA filter, the +/- 15v power supply is separate.
I much prefer this way of working where no wires are in the air but solder. No way for this type of project to take a breadboard.
You understand my approach because this U-shaped sheet is used to bring together the connections, the 2 L&R preamps, the buffered stereo RIAA filter, the +/- 15v power supply is separate.
I much prefer this way of working where no wires are in the air but solder. No way for this type of project to take a breadboard.
Reply
Show more
1 Comment(s)
josephfeng 3 years ago
I played with these a half century ago. The best RIAA equalization was in the Dynaco PAS/PAM preamps that had 1% capacitors and 1% resistors in the feedback loop. The values were wrong for op amps with infinite gain, but were perfect for the 12AX7 with a stage gain of about 100 (therefore, not as good as the tube aged and lost gain). For my listening the best sound and lowest noise from my Shure V15-II was from a Motorola MC1303, but this had saturation distortion with my Empire 888SE and some of my records. The dynamic range problem was exacerbated by the low power supply voltage, and there was a little more head room with with a uA709 or uA741 (but with poorer SNR).
Back in those days, Telefunken 112AX7's had a reputation for the lowest noise. Nowadays, the criterion seems to be the most euphonious distortion.
Back in those days, Telefunken 112AX7's had a reputation for the lowest noise. Nowadays, the criterion seems to be the most euphonious distortion.
Reply
Show more
0 Comment(s)
Tony Casey 3 years ago
It is misleading to suggest that changing the NE5534 for something "better" will improve the noise performance to the extent that you will notice it.
With an average moving magnet cartridge (1k5+680mH), the performance of the NE5534 is only about 1.5dB worse than a theoretical "noiseless" amplifier with DIN A weighting. Most of the noise comes from the source impedance rather than the opamp (unless you cool the cartridge to close to absolute zero, but then it wouldn't work).
It would only be worthwhile if you use a "high" output moving coil cartridge with a low impedance.
With an average moving magnet cartridge (1k5+680mH), the performance of the NE5534 is only about 1.5dB worse than a theoretical "noiseless" amplifier with DIN A weighting. Most of the noise comes from the source impedance rather than the opamp (unless you cool the cartridge to close to absolute zero, but then it wouldn't work).
It would only be worthwhile if you use a "high" output moving coil cartridge with a low impedance.
Reply
Show more
0 Comment(s)
PicoPete 3 years ago
This is an interesting project and the prototype looks carefully designed. However, I have to say that there is little explanation of the project implementation: There's no clarification about the circuit diagram or parts (BOM) or layout as it was implemented by the author. Considering the circuit deals with extremely low audio levels layout is critical - but there's nothing mentioned about layout in the article, neither is there any mention of construction details. The datasheet for the THAT IC shows considerably more complex circuits than the authors' schematic so it would be both instructive and useful to provide more explanation of what and why the author implemented what appears to be a very much simplified version. Lastly how closely does the prototype come to the RIAA curve and what about cartridge loading effects; Did the author plot response, noise. I don't mean to sound overly critical and probably there are others that will simply take the schematic and "go to town" with it. But when I see a project presented this way it tends to send me off running.
Reply
Show more
0 Comment(s)
Chris-H 3 years ago
This circuit doesn't even come close to the RIAA curves. There are several elements to RIAA correction, and these are often best done separately.
My RIAA circuit uses a lowpass filter first for the HF part of the curve, then the bass boost and LF rolloff are done in a second stage. This has the advantage that surface noise is significantly reduced, and scratches don't sound excessively loud! The original of this was built using the TL074, for use in a disco mixer. If it's wanted for audiophile use, you can apply any of the exotic low-noise op-amps.
I didn't bother with an MM pre-amp stage, because I haven't found any cartridge that comes close to the V15 or MC35C for broadcast use!
My RIAA circuit uses a lowpass filter first for the HF part of the curve, then the bass boost and LF rolloff are done in a second stage. This has the advantage that surface noise is significantly reduced, and scratches don't sound excessively loud! The original of this was built using the TL074, for use in a disco mixer. If it's wanted for audiophile use, you can apply any of the exotic low-noise op-amps.
I didn't bother with an MM pre-amp stage, because I haven't found any cartridge that comes close to the V15 or MC35C for broadcast use!
Reply
Show more
1 Attachment(s)
0 Comment(s)
Updates from the author
ERRYSON 2 weeks ago
pcb-with-ne5532-only.JPG (261kb)
ERRYSON 2 years ago
pu-mag-47k.JPG (234kb)
ERRYSON 2 years ago
ERRYSON 3 years ago
ERRYSON 3 years ago
http://www.retronik.fr/revues/radio-plans/1984/RP441_08-1984.pdf