A Low Noise Balanced Input Moving Coil Preamp Using the ZTX851

Where we discuss new analog design ideas for Pro Audio and modern spins on vintage ones.
Hans
Posts: 73
Joined: Fri Jun 24, 2016 8:48 am

Re: A Low Noise Balanced Input Moving Coil Preamp Using the ZTX851

Post by Hans »

Wayne,

I was still using your very first design with 15 Volt and 1k collector resistors, giving a collector current of 7.5 mA.
Changing this into 11V with the same 1k resistors, dramatically increased the noise by almost 2 dB, because of the lower collector current to 3.5 mA.
This explains to a great deal why you measured 0.52nV/rtHz.
When changing Vref from 7.33 to 5.5 Volt and lowering 1K to 600 Ohm, noise went even below the original low level, with a collector current of 9.2mA.

More to come,
Hans
Hans
Posts: 73
Joined: Fri Jun 24, 2016 8:48 am

Re: A Low Noise Balanced Input Moving Coil Preamp Using the ZTX851

Post by Hans »

The problem with the OPA2277 is though that it can only supply 10mA, so at this position this amp does not seem the best choice, unless you decrease the current and take the 2dB noise penalty for granted.
The OPA1612 that you initially used, seems a better choice for that matter.

Hans
Hans
Posts: 73
Joined: Fri Jun 24, 2016 8:48 am

Re: A Low Noise Balanced Input Moving Coil Preamp Using the ZTX851

Post by Hans »

But the OPA1612 has a diff Rin of only 20K, creating trouble with the high ohmic voltage divider R13-R16.
So I switched to an OPA1642, seemingly a better alternative.

Hans
User avatar
mediatechnology
Posts: 5442
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: A Low Noise Balanced Input Moving Coil Preamp Using the ZTX851

Post by mediatechnology »

Hans -

I think somewhere way back in this thread we came to the conclusion that we (I?) were going to target the >10Ω range of Rsource leaving <10Ω carts to the world of transformers. That is one of the reasons I lowered Ic.

Thanks for sim'ing it back at the higher currents.

Higher currents are do-able if the thermal mass of the input pair is large using the nut I DIY'd earlier and the op amps can provide Ie (≈Ic).
With less thermal mass and higher currents there are some self-heating effects.

Zooming out, how does the lower Ic impair NF with Rsource >10Ω based on sims?
Our sweet spot is a DL-103.

Not sure what you mean about the 10 mA output limitation of the OPA2277.
You may have thought I was suggesting the OPA2277 for the front-end op amp.
It's only used on the servo (right-hand side of the sch) and its outputs drive 20KΩ injection resistors.
In normal operation its output current is nowhere near 10 mA.

I would sim the front-end op amp as the OPA1612 or NJM2068.
Both did very well.
I know the 2068 isn't cool but it seems to hit the sweet spot combination of input noise voltage and current.
The OPA1612 is proabably the best-performing I have in my collection of go-to parts.
User avatar
mediatechnology
Posts: 5442
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: A Low Noise Balanced Input Moving Coil Preamp Using the ZTX851

Post by mediatechnology »

Hans wrote: Mon Sep 17, 2018 9:07 am But the OPA1612 has a diff Rin of only 20K, creating trouble with the high ohmic voltage divider R13-R16.
So I switched to an OPA1642, seemingly a better alternative.

Hans
OK.

Just thinking out loud...
If that's an issue we might also be able to use the servo op amp outputs to directly-drive the non-inverting inputs of the front-end op amp.
Move the bias elements and integrator to the input side of the servo.
The non-inverting inputs are driven by a 0Ω source and the integrator has a passive pole at the input.
Hans
Posts: 73
Joined: Fri Jun 24, 2016 8:48 am

Re: A Low Noise Balanced Input Moving Coil Preamp Using the ZTX851

Post by Hans »

Wayne,

I finalized my Sims.
A few remarks:
1) I mentioned the need to go with Vref from 7.5V to 5.5V. Initially I overlooked R17 R18 from the Servo being in par. to R15 R16 , so we landed at exactly the same point.
2) I was not sure what OPA you used for the input pair. Since I could only find a reference to the OPA2277 in your circuit, I used this one and found out that it could not source enough current. The OPA1612 has no such limitation, but the 20 K diff input creates a problem with the high ohmic voltage divider R13-R18.
The Opa1642 has a Fet input and can supply +/-30mA, enough to allow an input voltage of 30mV rms with a 1K collector load.
3) I also used an Opa1642 for the servo. Voltage noise of 5.1nV/rtHz of this amp, is hardly adding any noise to the output.
4) You mentioned that input resistance should be at least 10 Ohm, so that's what I did, although nowadays there are MC carts on the market with lower resistances.

Here are the results with Rsource 10 Ohm.
Vnoise flat: 75.75nV from 20 Hz to 20 Khz,the equivalent of 0.54nV/rtHz
Vnoise after Riaa + Aweighting: 30.55nV from 20 Hz to 20 Khz, the equivalent of 0.22nV/rtHz

Results with Rsource 1 Ohm
Vnoise flat: 52.62nV from 20 Hz to 20 Khz,the equivalent of 0.37nV/rtHz
Vnoise after Riaa + Aweighting: 21.23nV from 20 Hz to 20 Khz, the equivalent of 0.15nV/rtHz

Hans

P.S. I will try to upload the Spice model
Hans
Posts: 73
Joined: Fri Jun 24, 2016 8:48 am

Re: A Low Noise Balanced Input Moving Coil Preamp Using the ZTX851

Post by Hans »

Wayne Circuit_1.jpg
Hans
Posts: 73
Joined: Fri Jun 24, 2016 8:48 am

Re: A Low Noise Balanced Input Moving Coil Preamp Using the ZTX851

Post by Hans »

For using the That1240, I made a subcircuit with four 9k resistors around a configurable OPA, the combination tuned to exactly the specified -104dBu noise.
U1 to U4 are also making use of this configurable OPA, now set as an Opa1642.
I also inserted a voltage source between one emitter to simulate a Vbe difference.
With 5mV added, output offset was still below 10uV.

Hans
Hans
Posts: 73
Joined: Fri Jun 24, 2016 8:48 am

Re: A Low Noise Balanced Input Moving Coil Preamp Using the ZTX851

Post by Hans »

I posted the set of noise figures with a typo.
The first set was for Rsource 10 Ohm, but the second set was for Rsource 1 Ohm and not the 10 Ohm that I mentioned.

Hans
User avatar
mediatechnology
Posts: 5442
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Re: A Low Noise Balanced Input Moving Coil Preamp Using the ZTX851

Post by mediatechnology »

Thanks for the data.
I posted the set of noise figures with a typo.
The first set was for Rsource 10 Ohm, but the second set was for Rsource 1 Ohm and not the 10 Ohm that I mentioned.
I went in and fixed that typo for you. You should be able to go back and edit any past post. I don't recall a time limit for editing. You're not locked out from doing that here.

From my quick calculations converting your data it looks like the flat Noise figure with a 10 Ohm Rsource is about 2.5 dB. Ein -140.2 dBu.
At 1Ω Rsource flat NF is 9.3 dB. Ein -143.4 dBu.

It's been so long since I looked at this circuit I had forgotten what Ic actually is. 5.5 mA nominal.

I still stand by the fact that I rebuilt the original Protoboard when I changed the servo, am using a different bench supply with magnetic hum and may have some stray resistance in Rg, Rsource and for measuring gain my attenuator. All these measurement errors could easily add up.

I think the best think at this point is to dial in the circuit some more, lay a PC out, and measure it in a steel box.
Post Reply