His Master's Noise: A Thoroughly Modern Tube Phono Preamp - Page 32

21st April 2010, 12:00 PM
SY On Hiatus Join Date: Oct 2002 Location: Chicagoland	His Master's Noise: A Thoroughly Modern Tube Phono Preamp After more than 25 years of faithful service, it seemed that it might be time to redo my phono system. After all, I like to think that I've picked up a few tricks in the intervening years... The old system consisted of a VPI HW17-II, a Linn Ittok LVII tonearm, and a Troika cartridge. The Troika was... Click here to read the entire article Last edited by Variac; 2nd April 2011 at 12:17 AM.

31st March 2017

jackinnj

diyAudio Member

Why, silly hare, are you fleeing from the fierce jaws of the lion now grown tame?

31st March 2017

lehmanhill

diyAudio Member

Dave,

First of all, I am very far from expert at these things. But a couple of thoughts.

As I understand it, signal transformers change bandwidth depending on trafo gain and the load impedance following the trafo. As you consider trading off input trafo gain against amplifier gain, make sure you understand the bandwidth and noise implications for the trafo.

In other words, for a non-expert like me, I worry about the unknown tails that a change in the design creates. It seems like turning down the volume control is low risk by comparison. SY designed a high gain preamp with good noise immunity, so you aren't paying a price for the high gain as designed.

However, having said that, I do go ahead and take those risks, even when it's ill advised. Of course, I try to chase the tails all the way through the system, but I'm sure I don't understand every tradeoff. Maybe that's part of the fun.

Jac

31st March 2017

mrdave45

diyAudio Member

Hi jac, thanks for the response. I have looked at the loading info on the Rothwell site and as the 8055 has been superceded I would use the 1480. I would also build the test rig to check the loading. I worked out that around a minimum 3k should give suitable loading when the tx is set to 1:5. Without loss that gives me 2.8mV and with the loss due to the potential divider being set up with the cart impedance and the input load to give about 2.5mV on the output of the tx.
I haven't calculated the loss for the example cart. Maybe it's in the article. Can't remember off the top of my head but iirc about 1db down from 2mV.
I think bandwidth tends to be more of an issue with higher ratio transformers.
Maybe I'll build it stock with the 1480 at 1:10 with adjusted loading and see what happens. If nessesary I suppose I could add a potential divider between gain stages with some high grade resistors but I would have to research how do do that correctly. I think the impasse did something like that.

31st March 2017

kevinkr

diyAudio Moderator

I believe SY designed in more than sufficient headroom to accommodate your cartridge and the recommended (or replacement) SUT. A lower gain SUT such as the 8X LL1931 might be an option. I would not modify the circuit unless you understand the design intent since unpredicted consequences like incorrect EQ, increased noise and distortion are likely consequences of some of the changes you propose.

I generally design my phono stages to produce outputs with the intended cartridge and transformer in the range of typical red book CD players or DACs which is 2.0Vrms at 0dBfs. I think SY intended something similar in this design, but it has been a while since I last talked to him about it.

"To argue with a person who has renounced the use of reason is like administering medicine to the dead." - Thomas Paine

31st March 2017

mrdave45

diyAudio Member

Kevin, thanks, that's great to know.
I'll try it at 1:10
Wierdly my current atoc3 which is 0.35mV (not sure whether 3.54 or 5cms) in to a 55db preamp was only a bit quieter than my dac for the same cd/record, of course that's very non scientific as I know the dc isn't particularly hot and maybe the record was.
Still, I have the dilemma of whether to rewire my rb300 to balanced operation. I bought the better cartridge instead of upgrading to an audiomods at the same time. I'll probably leave as is until I can build the hmn then reasess.

7th April 2017

jackinnj

diyAudio Member

Some snapshots -- there are some power supply issues which need attention, otherwise, with EH 6922 the THD% is spot on:

HMN_THD_FFT.png

HMN_IMD.png

HMN_1K.png

HMN_50Hz.png

Why, silly hare, are you fleeing from the fierce jaws of the lion now grown tame?

7th April 2017

jackinnj

diyAudio Member

The PCB helicopter view:

HMN_PCB.jpg

Why, silly hare, are you fleeing from the fierce jaws of the lion now grown tame?

7th April 2017

mrdave45

diyAudio Member

Nice work! What's the 19k and 20k peak on the graph?

7th April 2017

jackinnj

diyAudio Member

Intermodulation Distortion. The spurs are almost 100dB below the mixed fundamental which is darn'd good.

Why, silly hare, are you fleeing from the fierce jaws of the lion now grown tame?

8th April 2017

mrdave45

diyAudio Member

100db anything seems pretty good to me! I wish my scope was that quiet. Think i need to get a better one.

So is your input signal a simulation before transformer at about ~0.2mV or post transformer straight in to the input on the board at ~2mV? 100db down from that seems insanely low noise, amazed thats even possible.