AudioRoundTable.com: Thermionic Emissions

Home » Audio » Thermionic Emissions » A 11 year long quest

Show: Today's Messages :: Navigator :: Polls
E-mail to friend

A 11 year long quest [message #94518]

Wed, 29 September 2021 21:51

gofar99
Messages: 2012
Registered: May 2010
Location: Southern Arizona

Illuminati (5th Degree)

Hi Everyone, Ever since the first tube based phono preamp I have wanted them to be quieter. I'll admit that the first few were nice sounding, but hummy and noisy. Later ones were really quiet but I felt that S/N of -75dbv was not the limit. Gradual improvements along the way brought it close to -80. So now I revised the design and completely over hauled the power supply as it seemed to be the thing holding me back. It was. Now with a different transformer, a medical grade SMPS for the heaters (neither over $20US, final three stages of filtering with poly caps and a capacitive multiplier in place of the regulators it is really, really quiet. I have reached the level where the tubes themselves are the biggest source of noise. They are selected EH7025s. With the better pair I show on the scope -93dbv, the worse pair -90. (see the image). The peak in the image is an artifact of the scope. Response is within +0.5db/-0.5db from 20HZ to 20K. The preamp has the same lovely sound as all the previous ones as that part is not changed. I'll post a schematic and photo of the guts in a day or so. In my system the listening level (on the preamps digital display) is right at -40db. When I increase it to 00 I need to have my ear within 6 inches of the speakers to hear anything. 40dbv is a ratio or 100 to 1. That is a lot of headroom!

/forum/index.php?t=getfile&id=2859&private=0

Attachment: Better channel noise2.png
(Size: 6.18KB, Downloaded 844 times)

Good Listening
Bruce

Report message to a moderator

Re: A 11 year long quest [message #94520 is a reply to message #94518]

Thu, 30 September 2021 09:30

Wayne Parham
Messages: 18978
Registered: January 2001

Illuminati (33rd Degree)

Very nice!

That's so important with high-efficiency speakers. You can hear any hint of noise on an efficient speaker, so to have that dead-quiet in between passages is really hard to find and is only realized in the best amplifiers.

Report message to a moderator

Re: A 11 year long quest [message #94538 is a reply to message #94520]

Fri, 01 October 2021 11:24

Rusty
Messages: 1415
Registered: May 2018
Location: Kansas City Missouri

Illuminati (3rd Degree)

I've come to realize that too. My new tubed line preamp with phono I hooked up to my old Hafler DH220 amp. And I can hear it's slight hum on the phono end. But using my tubed amplifiers it's smooth as baby's feet. Pi speakers let it all through. Like a spaghetti western know as The good, bad and ugly. The good is most music recorded, the bad is some and the ugly, is just that.

Report message to a moderator

Re: A 11 year long quest [message #96007 is a reply to message #94518]

Sat, 17 September 2022 23:20

positron
Messages: 145
Registered: May 2020

Master

Bruce, I was wondering if the +/- 0,5db was from the RIAA, the
rest of the circuit, or a combination of both. (I don't have the
schematic.)

For public.

Stanley Lipshitz presented an AES paper sometime back. He
covered both passive and active RIAA types. According to the
article, some sources/manufacturers are up to 7% off. If
anyone is interested, here is a link/calculator based on his
equations. (I found my simulator was off by several tenths of
a percent.)

http://www.mh-audio.nl/Calculators/CalculateRIAA.html

I would make sure the coupling capacitor next to V1 is plenty
large in value as its reactance "adds" to R1.

Just a thought that might help the public.

pos

Report message to a moderator

Re: A 11 year long quest [message #96009 is a reply to message #96007]

Tue, 20 September 2022 13:07

gofar99
Messages: 2012
Registered: May 2010
Location: Southern Arizona

Illuminati (5th Degree)

Hi, Yes indeed some of the formulas are off. I had to tweak the values to get the response I wanted. Getting RIAA accurate is not a simple task. This is what I do.

I use a general purpose calculator like the nl you listed (BTW that is a good site for many calculations) one to get some close values. Then comes the tedious part. I test the whole preamp not just the eq part. This allows for anomalies in the circuitry as well as the EQ. I use two different PC based scopes and a passive inverse RIAA like in the below schematic. One of the scopes does Bode plots. I checked its signal generator with the second scope to verify that it was not a source of deviation. The test and verify process indicated that the max deviation between 10HZ and 100KHZ was less than 0.1db with an input signal of 1.0 volts. Feeding this into the inverse EQ is where things can get a bit fuzzy. I have to accept that when built as indicated and fed a proper low Z input the results will be within about +/- 0.1db. To error on the cautious side I automatically presume it to be +/- 0.25 db. Then with that fed into the preamp (47K / 100pf in the cables no additional cap in the preamp) I run the plot. With tweaked values in the internal EQ it comes out well below +/- 0.5 db 20HZ to 20KHZ. Usually the values are in the negative 0.3db range. Either the values on the input side when added to the preamp tend to cancel (maybe) or it is spot on. Most of the plots are virtual straight lines with a tiny deviation on the negative side at the bottom end. So again to be safe I just tell everyone that it is +/- 0.5db. That value is low enough to be swamped by quirks in the cartridges. My experience has been that regardless of cost of the cartridge (mine go from $100 to over $1500) the linearity is worse than the 0.5db range.
/forum/index.php?t=getfile&id=3265&private=0