Go Back   Home > Forums > >
Home Forums Rules Articles diyAudio Store Blogs Gallery Wiki Register Donations FAQ Calendar Search Today's Posts Mark Forums Read

Analogue Source Turntables, Tonearms, Cartridges, Phono Stages, Tuners, Tape Recorders, etc.

Richard Lee's Ultra low Noise MC Head Amp
Richard Lee's Ultra low Noise MC Head Amp
Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 21st October 2019, 01:11 PM   #1651
Chris Hornbeck is offline Chris Hornbeck  United States
diyAudio Member
 
Chris Hornbeck's Avatar
 
Join Date: May 2011
Location: Little Rock
If they can't do it with a type ECC83, they just haven't tried hard enough. Harumpf. And you kids stop mumbling.


And furthermore,
__________________
"A marveilous newtrality have these things mathematicall and also a strange participation between things supernaturall, immortall, intellectuall, simple and indivisible, and things naturall, mortall, sensible, componded and divisible." - John Dee
  Reply With Quote
Old 21st October 2019, 01:13 PM   #1652
billshurv is online now billshurv  United Kingdom
diyAudio Member
 
billshurv's Avatar
 
Join Date: Mar 2014
Richard Lee's Ultra low Noise MC Head Amp
That does give me an idea...
  Reply With Quote
Old 21st October 2019, 02:10 PM   #1653
Hans Polak is online now Hans Polak  Netherlands
diyAudio Member
 
Join Date: Mar 2005
Location: Blaricum
Quote:
Originally Posted by syn08 View Post
I have to admit removing mains harmonics was always a challenging task for me. Admittedly, my lab is 6ft. away from the house 200 amp electrical panel, but I still don't understand how some people (you included) get these absolutely clean spectra? Not even a few nV which would be clearly visible in the spectra after x1000 gain?

I understand balanced connections, but your measurement preamp is single ended, how was it shielded/configured so that again not even a few nV of mains harmonics are leaking in? 10nV of 50Hz at the shorted input (-160dB) falls in a single bin and would be clearly visible against an -190dB noise floor. Mine is shielded in 12mm of aluminum and 5nV input equivalent 50Hz is about the best I can get (see #1609, 50Hz is -165dB equivalent at the input)
In my personal experience loops and incorrect connections of the mains gnd are the most frequent cause of mains pollution.

1) When first testing everything, isolated from my audio set, just Head Amp (HA) and 60dB measurement amp (MA), set up was as follows:
The 60dB SE MA was battery powered and driven from the HA's SE output.
The MA's output went to Scope+PC, the latter being connected to mains gnd.
(The MA powered from a wall wart gave lots of mains harmonics in the noise spectrum, most likely because the MA is not differential. That made the choice for batteries an easy one.)

So in effect, starting from the PC and through the Scope all signal grounds became connected to mains gnd with no possible gnd loop.
However when trying to additionally connecting the HA's signal gnd to mains gnd, the mains harmonics popped almost out of the screen, confirming the damage a gnd loop can do.
With the used setup, spectra from the 34dB Gain HA where clean with a 1R input resistor simulating the cart, for the Diff Current as well as for the Diff Voltage modules.
The spectra were clean with the Power Bank Battery and also with 2 different 5Watt USB mains adapters, and the Head Amp was not even encapsulated in an enclosure !
However with a 10Watt mains adapter several mains harmonics became visible.

2) The next step whith HA tested as part of the Audio set, this time being differential/balanced from start to end and with the cart connected, I was already quite happy with the results, bus as shown a few postings before, I had forgotten to connect a cable shield that even completely removed the remaining 50Hz component after connection.
For convenience to keep things universal, I have used RCA interconnects between Cart/ HA and HA/MC phono amp, but in both cases with a shield that is detached from the connector, with separate wires to be connected to an enclosure.
See the first and second image below.

Some CMRR measurement results all three made from 20Hz to 20Khz.
- The HA's measured CMRR with closely selected components is above 50dB
- The third image below shows the CMRR of my diff MC phono amp in combination with the preamp behind.
- The Main Amps have a measured >90dB CMRR in the Audio band.
If you want some more details, I'll be happy to provide them.


Hans
.
P.S. Just as additional info: When not connecting the HA's signal gnd to PE makes no difference at all when using the Power Bank, but with the USB mains adapter mains harmonics are all over the place.
So the PE completely neutralises these mains harmonics.
Attached Images
File Type: jpg CONNECTION.jpg (220.9 KB, 249 views)
File Type: jpg Interconnect.jpg (250.3 KB, 220 views)
File Type: jpg CMRR MC+Pre.jpg (103.2 KB, 253 views)

Last edited by Hans Polak; 21st October 2019 at 02:22 PM.
  Reply With Quote
Old 21st October 2019, 02:17 PM   #1654
Hans Polak is online now Hans Polak  Netherlands
diyAudio Member
 
Join Date: Mar 2005
Location: Blaricum
Quote:
Originally Posted by billshurv View Post
I should note that at some point the cartridge itself will be picking up more mains hum than everything else, so we pretty much have a hard limit unless the whole front end is in a seperate metal box with no mains!
That's a valid remark.
I have only measured the noise spectrum in the complete Audio set with the Current modules that present a short circuit to the Cart.
Could be that the Voltage modules with Rin is 260R make picking up mains a bit more susceptible.
I will test this and report back the results.


Hans
  Reply With Quote
Old 21st October 2019, 02:44 PM   #1655
syn08 is offline syn08  Canada
diyAudio Member
 
syn08's Avatar
 
Join Date: Aug 2005
Location: Toronto
Quote:
Originally Posted by Hans Polak View Post
In my personal experience loops and incorrect connections of the mains gnd are the most frequent cause of mains pollution.

1) When first testing everything, isolated from my audio set, just Head Amp (HA) and 60dB measurement amp (MA), set up was as follows:
The 60dB SE MA was battery powered and driven from the HA's SE output.
The MA's output went to Scope+PC, the latter being connected to mains gnd.
(The MA powered from a wall wart gave lots of mains harmonics in the noise spectrum, most likely because the MA is not differential. That made the choice for batteries an easy one.)
So correct me if I'm wrong, for measurement purposes both the HA and the MA were battery powered, correct? That would explain it, but then how come your comment in #1632 and #1633 (with the clean spectra, that is still a measurement) that USB bank or wall wart makes no difference? That's what I find strange, even more so you are using a PC, not really famous for low noise environment. And no shielding for the HA, almost unbelievable! After a total gain of 34dB+60dB, not even a microvolt of mains harmonic at the output, meaning less than 20pV equivalent at the input. There is definitely more than "no ground loops" here.

And you are not alone, I've seen the same with other results, it is frustrating. I'm fighting any quarter of nanovolt contamination and others have absolutely no problem in getting absolutely clean results. OTOH, I also see in professional papers measurements affected by the mains harmonics, to the point I started to take this harmonic contamination as a proof of results credibility.

Speaking about PCs, I recently acquired a brand new 3rd Generation Focusrite 2i2 interface, USB powered. Excellent performance, but even with the balanced input shorted at the input jacks I can still see crud (mains harmonics and other) creeping in the output spectra. Once again, not a big deal, but the same question about perfectly clean spectra you posted arises.

How did you cleanly measure 120dB of PSRR? I'm aware of the usual PSRR measurement techniques, but in practice that's another thing that I was never able to do.
  Reply With Quote
Old 21st October 2019, 05:31 PM   #1656
sgrossklass is offline sgrossklass  Germany
diyAudio Member
 
Join Date: Nov 2004
Location: Germany
Quote:
Originally Posted by Hans Polak View Post
P.S. Just as additional info: When not connecting the HA's signal gnd to PE makes no difference at all when using the Power Bank, but with the USB mains adapter mains harmonics are all over the place.
So the PE completely neutralises these mains harmonics.
And I think I know why that is.

Wall wart style 2-prong SMPS like your USB power adapter have a tendency to connect the mains filter caps (typically 2x2n2 or somesuch) to secondary-side ground. So your head amp ground is being pulled towards 1/2 * 230 V~, and now the poor MC amp's input CMRR has to do all the heavy lifting since the cable shield is not hooked up. Not to mention potential capacitive coupling to the head amp's input end.

I bet you could connect the cable shield at the MC amp end and the mains harmonics would disappear entirely, or close to it anyway. A dedicated PE connection still is lower-impedance and with the shield not having to carry any current I would assume your usual setup to be higher performance, but whether you would actually see the difference, I don't know. If this measurement setup doesn't show it I don't know what would!

BTW, regarding the electrolytic capacitor adventures posted here earlier - did you know that it has actually been shown that capacitors can be characterized for leakage by 1/f noise, i.e. the latter being a side-effect of the former? Very low ESR is generally connected to high leakage, so the findings make complete sense. One to be aware of when you're relying on RC filtering for PSRR or have high-impedance nodes tied to caps.

Leakage is also tied to longevity (where ripple is not an issue). The orange low leage ELNAs in the MPX section of tuners are generally still good after 4 decades despite never seeing much of any DC, regular 10µ/16V polars kept under similar conditions are generally toast.

Last edited by sgrossklass; 21st October 2019 at 05:44 PM.
  Reply With Quote
Old 22nd October 2019, 10:03 AM   #1657
billshurv is online now billshurv  United Kingdom
diyAudio Member
 
billshurv's Avatar
 
Join Date: Mar 2014
Richard Lee's Ultra low Noise MC Head Amp
Quote:
Originally Posted by Hans Polak View Post
. But as shown in the above spectra, with the PE connected to the Head Amp’s analog gnd, in this very case the noise spectrum is as clean as with a battery (Power Bank)

Hans

I just noted this. I had assumed only case connected to PE as the power supply is effectively floating but for the wall wart case that makes sense.
  Reply With Quote
Old 22nd October 2019, 02:06 PM   #1658
Hans Polak is online now Hans Polak  Netherlands
diyAudio Member
 
Join Date: Mar 2005
Location: Blaricum
Quote:
Originally Posted by syn08 View Post
So correct me if I'm wrong, for measurement purposes both the HA and the MA were battery powered, correct?
MA was always powered by a battery but the HA was not just powered by a Power Bank in the test setup but also by several mains adapters.
I mentioned for the Test Setup with the MA + HA that “The spectra were clean with the Power Bank Battery and also with 2 different 5Watt USB mains adapters, and the Head Amp was not even encapsulated in an enclosure !
However with a 10Watt mains adapter several mains harmonics became visible. “.
Quote:
but then how come your comment in #1632 and #1633 (with the clean spectra, that is still a measurement) that USB bank or wall wart makes no difference? That's what I find strange, even more so you are using a PC, not really famous for low noise environment. And no shielding for the HA, almost unbelievable! After a total gain of 34dB+60dB, not even a microvolt of mains harmonic at the output, meaning less than 20pV equivalent at the input. There is definitely more than "no ground loops" here.

And you are not alone, I've seen the same with other results, it is frustrating. I'm fighting any quarter of nanovolt contamination and others have absolutely no problem in getting absolutely clean results. OTOH, I also see in professional papers measurements affected by the mains harmonics, to the point I started to take this harmonic contamination as a proof of results credibility.
The images in #1632 and #1633 that you refer to, were measured spectra with the HA as part of a complete Audio set with Cart connected. So in this case no MA in use but a fully differential in/out MC Riaa phono preamp instead.

I fail to understand your 20pV calculation. Level of the shown spectra are at 312pV/rtHz, so mains harmonics are less than 312pV and not 20pV.

Quote:
Speaking about PCs, I recently acquired a brand new 3rd Generation Focusrite 2i2 interface, USB powered. Excellent performance, but even with the balanced input shorted at the input jacks I can still see crud (mains harmonics and other) creeping in the output spectra. Once again, not a big deal, but the same question about perfectly clean spectra you posted arises.
Because the 2i2 is powered from the PC’s noisy supply through the USB interface, it will be susceptible to inflow of mains harmonics. The 2i2 is like a Diff HA and a SE MA in one box, were my MA had to be powered by batteries.
My Single Ended MA absolutely needed a battery supply to achieve a clean spectrum, as opposed to the diff MC Phono amp with its own separate PE connection that didn’t have to be supplied from batteries to get a clean spectrum.
So could it be that fully differential in/out, in combination with its own separate local connection to PE is the all-important reason why? For the time being, to my experience the answer is a clear yes.
A little bit diff as in the 2i2 does not work IMO. It should be either diff from start to end with a separate PE connection or alternatively SE, powered by batteries for the sensitive parts.

In case of my differential HA, the USB mains adapter has no direct connection to the signal gnd, and that, in combo with the diff topology, might be the reason why I can use a wall wart as long as the signal gnd is connected to PE, either from the PC in the measurement setup, or with its own PE connection in the fully differential Audio chain.


Quote:
How did you cleanly measure 120dB of PSRR? I'm aware of the usual PSRR measurement techniques, but in practice that's another thing that I was never able to do.
I did not mention PSRR but CMRR.
Testing the CMRR was done the “simple way” by connecting a signal generator’s signal between the short-circuited inputs and signal gnd.
Measured results do conform perfectly well to fig 12 in the AD797’s specs.
Scott did an excellent job with the design of this op-amp.

Hans
  Reply With Quote
Old 22nd October 2019, 02:42 PM   #1659
syn08 is offline syn08  Canada
diyAudio Member
 
syn08's Avatar
 
Join Date: Aug 2005
Location: Toronto
Quote:
Originally Posted by Hans Polak View Post
MA was always powered by a battery but the HA was not just powered by a Power Bank in the test setup but also by several mains adapters.
I mentioned for the Test Setup with the MA + HA that “The spectra were clean with the Power Bank Battery and also with 2 different 5Watt USB mains adapters, and the Head Amp was not even encapsulated in an enclosure !
However with a 10Watt mains adapter several mains harmonics became visible. “.
That's exactly what I don't understand:

- Single ended measurement setup: HA+MA+PC
- HA powered from 5W USB mains adapter (alleged switching)
- MA powered by batteries
- HA not shielded
- Total gain, about 96dB (36dB HA + 60dB MA)
- Cables used, connectors, unknown (type, length)

Result: less than 312pV (input equivalent) of mains harmonics at the output (20pV was considered as excess over the 312pV/rtHz "floor" to make it visible on the spectra).

I'll post later the effect of a 30cm single ended cable (shorted) at my MA input. This effect has nothing to do with ground loops, etc...
  Reply With Quote
Old 22nd October 2019, 04:36 PM   #1660
Hans Polak is online now Hans Polak  Netherlands
diyAudio Member
 
Join Date: Mar 2005
Location: Blaricum
Quote:
Originally Posted by syn08 View Post
That's exactly what I don't understand:

- Single ended measurement setup: HA+MA+PC
- HA powered from 5W USB mains adapter (alleged switching)
- MA powered by batteries
- HA not shielded
- Total gain, about 96dB (36dB HA + 60dB MA)
- Cables used, connectors, unknown (type, length)

Result: less than 312pV (input equivalent) of mains harmonics at the output (20pV was considered as excess over the 312pV/rtHz "floor" to make it visible on the spectra).

I'll post later the effect of a 30cm single ended cable (shorted) at my MA input. This effect has nothing to do with ground loops, etc...
Correct for the above, with the following additions:

-Unshielded HA on mouse pad, powered from 5W USB adapter having no direct connection with signal gnd (called Null here), see second image below.
-HA diff input terminated with 1R resistor, no cable connected at this side
-connection between HA and MA with 60cm Siltech FTM4 Gold cable with WBT RCA connectors, driven from HA's Diff to SE converter.
-Connection between MA and scope a 60cm 50R coax cable.

Picture is still with the Nichicons that were later replaced by Panasonics FM.


Hans
.
Attached Images
File Type: jpg Head Amp.jpg (630.7 KB, 143 views)
File Type: jpg Mother Board.jpg (279.3 KB, 147 views)
  Reply With Quote

Reply


Richard Lee's Ultra low Noise MC Head AmpHide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
ultra low noise regulators wtnsmk Power Supplies 26 2nd May 2019 04:12 PM
ultra low noise power supply and low noise amplifier ambrosia168 Power Supplies 8 13th October 2017 03:43 PM
Matching a 1970's receiver with these Richard Allan Loudspeakers? (Richard Allen) gearu Full Range 9 23rd September 2014 04:36 AM
Ultra Low Noise - Ultra Low Impedance - Ultra big Problems! Stefan0815 Analog Line Level 2 14th January 2014 07:01 AM
Ultra low noise filters tiroth Solid State 14 21st May 2002 11:19 PM


New To Site? Need Help?

All times are GMT. The time now is 05:09 PM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
Copyright ©1999-2019 diyAudio
Wiki