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.
billshurv
Posts: 436
Joined: Thu Feb 01, 2018 6:07 am

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

Post by billshurv »

Hans: I assume you have the ability to measure this stuff (which I don't). I have 2 boards from Wayne at the moment that I will be soldering up soon, so happy to send one to you for tests as I've got a heap of MM experimentation to get out the way so won't get around to using that until probably christmas!
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 »

Bill,
What a great idea.
I will be happy to do the measurements.

Hans
billshurv
Posts: 436
Joined: Thu Feb 01, 2018 6:07 am

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

Post by billshurv »

Encourages me to break out the spare Round Tuit :)
User avatar
JR.
Posts: 3697
Joined: Sat May 24, 2008 7:21 pm
Contact:

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

Post by JR. »

I have a round tuit but I don't know how to directly measure Rbb, I guess you can infer it from ein voltage.

JR

[edit- BUT I remember seeing an artifact caused by Rbb that may reveal a way to measure it.

Back in the 1980s my Loftech TS-1 performed a log conversion using small geometry bipolar transistors inside a common array. I was able to measure an error in the dB reading at higher base-emitter currents. The TS-1 read a major fraction of 1 dB higher at +20dBu due to Rbb's IxR voltage drop in addition to the base-emitter junction voltage.

To do this perform a high precision base emitter voltage vs current measurement, at high enough current for the Rbb to show up in the voltage measurement. The TS-1 expressed errors at low single digit mA, a modern low Rbb device would likely need higher current than that to reveal its Rbb contribution.

Note: this measurement would need to be normalized for temperature since that will also affect Vbe. Perhaps servo the temperature by varying the collector voltage to manage internal power dissipation.

I didn't say this would be easy.... :lol:

[/edit]
Cancel the "cancel culture", do not support mob hatred.
User avatar
mediatechnology
Posts: 5437
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 -

Looking back at the sim you posted earlier here I was wondering how the 1240 was modeled.
Does the 1240 model in your sim realize common mode rejection?
Is it a true diff amp model?

I ask because with "high" Rbias resistors (499Ω x2) and a low source R connected differentially (10-40Ω typ.) a significant portion of each device's noise current is shunted between the inputs, appears in common mode, and is rejected by the 1240s.
If the 1240 models for some reason don't realize common mode rejection, then the sim and circuit will be different.

I'm not sure if it matters too much if rbb' is 1.5Ω or 2Ω because the ZTX-devices we can get are about as good as its gonna get.
My estimation of rbb' is based on inferring it from the output noise.

I have a paper on how to measure it but IIRC it was crazy-complicated.
User avatar
JR.
Posts: 3697
Joined: Sat May 24, 2008 7:21 pm
Contact:

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

Post by JR. »

mediatechnology wrote: Mon Jul 01, 2019 1:25 pm

I'm not sure if it matters too much if rbb' is 1.5Ω or 2Ω because the ZTX-devices we can get are about as good as its gonna get.
My estimation of rbb' is based on inferring it from the output noise.

I have a paper on how to measure it but IIRC it was crazy-complicated.
agreed... when working at the bleeding edge of low noise technology its just about being better than the other part. The best confirmation of that is in an actual circuit. I was just trying to answer the question asked.

There may be high current power transistors with lower Rbb but probably not lower noise in circuit.

JR
Cancel the "cancel culture", do not support mob hatred.
User avatar
mediatechnology
Posts: 5437
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 read back through the thread where you described how you modeled the THAT1240s and they are correct in that they should realize CM rejection.

I'm not sure which set of measurements that you're referring to but there's always a possibility of an unaccounted for 0.7 dB error due to instrument loading.
The PCM4222 loads a 50Ω (THAT1646) or 100Ω (THAT1240 X2 with 49Ω build out) in about 0.7 dB.

In most of my measurements you'll see the approximate 0.7dB accounted for but there may be one where it isn't making the Ein measurement 0.7dB better.
That error would make a 540 pV noise density look like 500 pV.

This measurement here is on the PC board: viewtopic.php?f=6&t=783&start=380#p13062
mediatechnology wrote: Sat Apr 13, 2019 2:35 pm Image
ZTX851 Moving Coil Preamp Output Noise 62dB Gain

Rsource is 10Ω.
The output noise floor is -80.19 dBu indicated.
There is 0.74 dB loading by the A/D so the actual output level, unweighted, is -79.45 dBu.
The measured gain is +61.61 dB.
Thus, the Ein is -141.06 dBu.

The Rnv is 14.5Ω.
Subtracting Rsouce+Rgain from Rnv leaves 3.5Ω.
Thus rbb' is estimated to be around 1.75Ω per transistor.

The NF for a 10Ω source is 1.6 dB.
By my quick calculations it works out to be 0.484nV√Hz and does reflect the measurement loading.
IIRC the measurement BW is 20 kHz (as defined by AudioTester) though the display range is 22 kHz.
Not sure what AudioTester's 20 kHz measurement bandwidth works out to as equivalent noise BW.
User avatar
mediatechnology
Posts: 5437
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

MC Preamp vs DL-103 Hum Signatures

Post by mediatechnology »

I thought I would grab some quick shots of the low-level hum performance of the preamp and the "normal," differential mode, hum pickup of the DL-103.

The preamp wiring is STP/balanced/floating to the head shell.

The first image is an FFT with the DL-103/head shell unplugged from the arm to get a background noise level of the preamp with all tone arm wiring connected. The noise level in the first frame is determined by the internal 100Ω termination.

Image
ZTX851 Moving Coil Preamp Hum Signature Headshell Removed, Flat, Gain=58dB.

The next image is with the DL-103 and headshell plugged into the arm with the arm on the rest.

The hum shown here is the 1-2 mG ambient field inducing differential mode hum into the DL-103's coils. The noise floor lowers about 4 dB because the 40Ω cart is now in parallel with the 100Ω termination. (Note that common mode rejection cannot reduce differential mode hum in the source.)

Image
ZTX851 Moving Coil Preamp Hum Signature DL-103 on Armrest, Flat, Gain=58dB.

As the arm is moved onto the platter the DL-103's orientation to the field can be seen to change:

Image
ZTX851 Moving Coil Preamp Hum Signature DL-103 Over Platter, Flat, Gain=58dB.

The DL-103's hum is probably about normal and is inaudible under surface noise.
In an area with a lower ambient hum field it might be lower.
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 »

mediatechnology wrote: Mon Jul 01, 2019 1:25 pm Hans -

Looking back at the sim you posted earlier here I was wondering how the 1240 was modeled.
Does the 1240 model in your sim realize common mode rejection?
Is it a true diff amp model?

I ask because with "high" Rbias resistors (499Ω x2) and a low source R connected differentially (10-40Ω typ.) a significant portion of each device's noise current is shunted between the inputs, appears in common mode, and is rejected by the 1240s.
If the 1240 models for some reason don't realize common mode rejection, then the sim and circuit will be different.

I'm not sure if it matters too much if rbb' is 1.5Ω or 2Ω because the ZTX-devices we can get are about as good as its gonna get.
My estimation of rbb' is based on inferring it from the output noise.

I have a paper on how to measure it but IIRC it was crazy-complicated.
Hi Wayne,

Below the image showing the measured noise up to 22kHz, measuring 4.86uV or -104dBu conform the specs.
Common mode rejection is even far better as the real one.

Apart from that, when measuring noise before the That's at the outputs of the NE5532, difference RTI is only 0.02dB, so the contribution of both That's is neglectable.

Hans
Wayne Circuit_3.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 »

Wayne,

It's always nice to compare things.
Image below shows noise with Cart on the Platter with motor off, resp with motor on but belt removed recorded with my Diff in Diff Out preamp.
Image was originally made after Riaa correction, but with Audacity an inverse Riaa was applied, so noise above 10kHz is unreliable and should be flat.

Hans
Motor-on-off.jpg
Post Reply