Flat Phono Preamp Based on John's P10 and 2SK389

Where we discuss new analog design ideas for Pro Audio and modern spins on vintage ones.
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Flat Balanced Input/Balanced Output Moving Magnet Phono Prea

Post by mediatechnology »

This is the latest schematic of the Flat Balanced Input/Balanced Output Moving Magnet Phono Preamp.
The preamp can be located remotely at the turntable and/or lathe.
Another potential application for this board is a cutter head feedback coil preamp.

I'll have boards to test in a couple of days.

Image
A Flat Balanced Input/Balanced Output Moving Magnet Phono Preamp

The next step is the RIAA EQ/Monitor Switcher...
Didn't I send you one of my gold 4 pos dip switches from my cap loading board kit?
I remember the boards and will look for the DIP switches.
I decided to use jumpers on this one.
Thought I might be able to pack it tighter.
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Flat Phono Preamp Based on John's P10 and 2SK389

Post by mediatechnology »

The board:

Image
A flat, non-EQ'd, balanced input, balanced output phono preamplifier for RAW transfers using DSP-based de-clicking and DSP RIAA EQ.

The initial run of four boards were virtually perfect.
The only thing I found to change was to slide the "Power" silkscreen over so the "P" doesn't land on a feethrough.

Image
A Flat Balanced Input/Balanced Output Moving Magnet Phono Preamp

I split the gain stop resistors in the INA to reduce the potential reduction of HF CMR from pin capacitance with off-board gain switching.

I'm very happy with the NJM2114 dual op amps.
I have some JRC2068DD op amps coming in tomorrow to also try.
The "DD" "double-D" is the selected for low noise RIAA-weighted parts.
Now that this is off the protoboard I'll try more of the LME49720/LME49860.

If this board is to be used to directly drive a soundcard or A/D I would install C4/C5/C24/C25.
I left mine completely DC-coupled since the RIAA/Monitor switcher will have a 1 uF PP downstream film cap.
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: Flat Phono Preamp Based on John's P10 and 2SK389

Post by JR. »

It seems a little odd that this late in the life cycle of vinyl we can buy op amps selected for RIAA noise, while back in the day about the only dedicated parts designed for vinyl playback was National Semi's LM381/387 parts.

JR
Cancel the "cancel culture", do not support mob hatred.
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Noise Test Using The NJM2114, 2068, 5532, 33078 and LME49720

Post by mediatechnology »

I just compared the noise performance of the NJM2114, NJM2068, NJM5532, MC33078, LME49720 and LME49860.

And the winner is: The NJM2068DD "double-D" "D-rank" grade.
(Measured with a 2K Rsource at 40 dB gain. I decided not to use a dummy cart for this test.)

The NJM2144 is a close second.
The 5532 and MC33078 are tied at third.

The LME49720 and LME49860 are unuseable due to burst noise. (See: https://e2e.ti.com/support/amplifiers/a ... 6/t/415907)

These are noise comparisons of using both channels.
This is the flat output of the preamp before RIAA EQ.

Image
Noise comparison of the Left and Right Channels using NJM2114s in both channels.

Image
Noise comparison of the Left and Right Channels. NJM2068 in Red, NJM2114 in Green

The NJM2068 had the lowest noise and was used as the basis for comparison.

Image
Noise comparison of the Left and Right Channels. NJM2068 in Red, NJM5532 in Green

Image
Noise comparison of the Left and Right Channels. NJM2068 in Red, MC33078 in Green

The LME49720 and LME49860 are a solid "fail."
I tried numerous samples from various date codes and could not find a sample suitable for test.
Both the LME49720 and LME49860 suffer from burst noise:

Image
LME49720 burst noise. Measured at INA output 40 dB gain. 5 mV/2ms per division.

In between bursts the LME49720 performs quite well - as good as the NJM2068.
TI needs to get a handle on this - I'm not the only one that has found it.
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: Flat Phono Preamp Based on John's P10 and 2SK389

Post by JR. »

I am struck by how well the technology has evolved to make pretty noise floor plots. Could I get one of those systems delivered to my bench circa 1980? :lol:

I had to expand the scale to read the differences and it looks like low single digit dB (I think?). While better is always better. Do your analyzers have any bandwidth/weighting capability? I had a fairly crude IHF "A" filter (with gain) that I used the hell out of back in the day because those results correlated well with my ears.

That shot noise looks unacceptable for an audio path, unless those are silly cheap and can only be used at low closed loop gain. A few tens of microvolts will not be a huge issue in a unity gain buffer, but again better is always better. I hope it's only a process issue and not the underlying technology, but luckily for me I can let you guys sort that out. 8-)

Nice stuff...

JR
Cancel the "cancel culture", do not support mob hatred.
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Flat Phono Preamp Based on John's P10 and 2SK389

Post by mediatechnology »

I had to expand the scale to read the differences and it looks like low single digit dB (I think?). While better is always better. Do your analyzers have any bandwidth/weighting capability? I had a fairly crude IHF "A" filter (with gain) that I used the hell out of back in the day because those results correlated well with my ears.
Yep I think it's about a dB quieter.
And the NJM2068 is the least-expensive too...

My analyzer has filters but not good ones.
The documentation isn't that good either.

I needed an analog 20 kHz noise filter yesterday: viewtopic.php?f=6&t=749
had to settle for a 100 kHz 6dB/octave and a 157 kHz noise bandwidth.
User avatar
JR.
Posts: 3700
Joined: Sat May 24, 2008 7:21 pm
Contact:

Re: Flat Phono Preamp Based on John's P10 and 2SK389

Post by JR. »

mediatechnology wrote:
I needed an analog 20 kHz noise filter yesterday: viewtopic.php?f=6&t=749
had to settle for a 100 kHz 6dB/octave and a 157 kHz noise bandwidth.
I don't recall the exact cut-off but IIRC Paul Buff wrote a pretty good article on the subject of noise measurements back in the '80s (in RE/P). He suggested you could approximate a 20kHz measurement bandwidth with a one pole LPF at around 12kHz. For measurement purposes the HF cut off shape/slope was a major source of error. Starting the one-pole rolloff enough below 20kHz, balanced out the below 20kHz noise dropped, with the above 20khz noise still left in the measurement. Using a passive LPF eliminated filter/amplifier noise as a complication.


JR
Cancel the "cancel culture", do not support mob hatred.
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: Flat Phono Preamp Based on John's P10 and 2SK389

Post by mediatechnology »

That's a good idea.

The factor for a single-order is filter is 1.5708 so an fc of 12.73 kHz would have an equivalent BW of 20 kHz.

Equivalent Noise Bandwidth Chart. viewtopic.php?f=12&t=475&p=5331
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Noise Measurements of the MM Preamp

Post by mediatechnology »

I finally had a chance to wire the preamp umbilical up properly.
I trimmed the shielded twisted pair balanced DIN tonearm leads to 3 feet.
The line output leads from the flat preamp are 48 feet feeding the RIAA EQ/Monitor protoboard.
I had NJM2068s already installed in the flat preamp so I used them.
It "sounds" quiet. Scary quiet.

I wanted to see what the noise level was when gain was set to produce +4dBu output.
The final output has 6 dB gain and the RIAA decoder was set to unity gain.
The remainder of the gain, all but 6 dB, was dialed into the preamp front end.

The ATR-96 MM cartridge is rated 5mv at 5 cm/s 1 kHz. They don't specify whether modulation is mono or single-channel.
The CBS STR-100 test record I'm using for level set is 3.54 cm/s with one channel modulated.
If I'm not mistaken this should work out to be 5 cm/s with both channels driven.
Based on that premise, I set the single channel gain to a reference level of +1 dBu.
Both channels driven should produce +4 dBu at 5 cm/s (lateral).
I may stand corrected on that...

The RIAA EQ'd (but not A-weighted) output noise with a real cart, sitting on a real turntable (not a resistor) produced an output noise floor of about -72 dBu.
This measurement bandwidth is 157 kHz. (fc=100 kHz.)
The signal to noise, unweighted, referenced to +4 dBu is about 76 dB.
Some of that is hum at Line and 2X, 3X and 5X line.

Its quiet, really quiet.
I suspect most of its quietness is from the balanced interface and common mode rejection.

BTW for someone wanting a FET input, the OPA2134 worked pretty good.
It was about 3 dB noisier (with a resistive input termination) than the NJM2068.
The OPA2134 may actually be better with the cart, compared to a resistor, due to its lower current noise.
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

RIAA EQ/Monitor Switcher Schematic

Post by mediatechnology »

After using and living with the flat phono preamp and RIAA EQ monitor switcher for awhile I think I know where to put all the jumpers.
Pauls' suggestions, particularly the inverse RIAA EQ for dub plate cutting, really helped.


I tried to put in a lot of flexibility and make it a "Swiss Army knife."
Someone wanting a simple RIAA EQ'd phono preamp needs only to build that part of the board.
My application, for LP archiving using RAW flat transfer, requires the full circuitry.

This is the schematic for the RIAA EQ and Monitor switcher:

Image
RIAA EQ, Inverse RIAA EQ and Monitor Switcher for the Flat Phono Preamp.

Signal Flow

The design has two signal paths. One is for recording the second for confidence monitoring or simultaneous recording and playback.
An RIAA EQ section can be placed in either the record path or playback path.

Flat transfers can be made through the record path while non-destructively switching monitor modes from monitoring flat record, record with EQ or playback from the DAW output.
Having analog RIAA EQ for monitoring saves ears and permits later de-clicking, done in the RAW un-EQ'd domain, to be heard in context.
Final EQ of the transfered files is done in DSP.

The flat balanced input/balanced output moving magnet preamp can be remotely located and powered from the monitor switcher permitting long cables between the turntable and workstation.

The flat/wideband preamp output is received by a THAT1240 line receiver.
It is then AC-coupled into a variable gain inverting buffer.
Relay RY2 selects either flat or RIAA EQ recording mode. RY1 must also be in Record.
The operation of RY1 and RY2 are destructive.
The record out is available as an RCA connection or fully-balanced by a THAT1646.

The RIAA EQ section can be switched between the Record and Playback Monitor paths by RY1.
A post-EQ gain stage can be used to provide stepped gain trim and additional post-EQ gain if needed.

The DAW playback output is fed into the THAT1246-based line receiver.
RY3 selects either Flat or EQ'd monitoring from either the Play line input, Preamp Output or RIAA EQ output.
The operation of RY3 is non-destructive to recording.

Jumper Description

J1 bypasses the balanced input stage. This is used when the flat preamp is in the same chassis as the EQ/Monitor Switcher.
When J1 is connected directly to the input, the THAT1240 on the EQ board and one THAT1240 on the flat preamp may be removed.

J2 (optional) permits the preamp gain to be raised or lowered in 3dB steps. It can also be a front panel switch.

J3 is a rumble/warp filter insert.

J4 is for alternate EQ. It can be used to bypass the RIAA stage and for inverse RIAA dub cutting allow the insert of a 50 kHz ultrasonic filter. (e.g. Neumann 2 pole SK)

J5A/J5B switch the RIAA filter to inverse RIAA to permit the board to be used for cutting discs.

J6 is an optional cartridge correction filter insert point. J6 may also be used to provide a return for alternate EQ.

J7 permits the post-EQ gain to be switched in 3 dB steps. J7 can also be used to sum cutterhead feedback with modulation.

J8 is a second playback cartridge correction filter insert. J8 can also be used for additional cutterhead EQ.

J9 selects the flat monitoring source. J9 can either be connected to the Playback line input or it can follow the RIAA EQ. J9 can also be made a switch. For transfers which are going to primarily be made without RIAA EQ, "RAW," then J9 should be jumpered to follow the RIAA EQ. If RIAA EQ'd transfers are the norm, placing the RIAA EQ in the record path, J9 should be jumpered to Play line in to permit confidence monitoring or simultaneous editing of files already recorded with EQ.

Board Stuffing Options

There are lots of them but these are the highlights.

The preamp gain can have an optional trim pot to adjust flat gain.
Gain is typically trimmed in the flat moving magnet preamp.
For moving coil preamps, fine adjustment of gain is not always possible.
A trim in this location allows minor amounts of "helper" preamp gain and adjustment of channel balance for moving coil carts.

If the flat preamp is located with the EQ/Monitor switcher the THAT1240 balanced input is not required.
The flat preamp can have one THAT1240 removed.
Due to signal inversion, the inverting output of the flat preamp and inverting input of the EQ board are used to maintain correct polarity.
This is also the reason the THAT1240 appears "upside down" with the + input connected to the - THAT1240 input.

The RIAA EQ shows optional capacitors bridging the 10 nF.
The 10 nF do not require them.
If precise 10 nF are not available, a combination of 6800 and 1500 pF can be stuffed.
The resistor values then become 11K8, 31K8 and 318K with links in the unused resistor locations.

For applications which do not require post-EQ gain trim it may be eliminated.

Workflow Enhancement

Two simultaneous signal paths allow flat transfers to take place while other files are de-clicked, edited and rendered with RIAA EQ in DSP.
Having two paths is a huge time saver for archival transfer because it allows to things to be done at once.
There will be no cigarette breaks for interns doing transfers with this box - if you got time to lean you got time to edit.
Post Reply