Noise from TT output jacks?

Jump to Latest
#43
In order to know what's important, you first have to know the theoretical limits.
This graph is for theoretically ideal resistors.
From https://www.rfcafe.com/references/electrical/noise-power.htm :

1703205151233.png
 
Last edited:
  • Like
Reactions: Chris Hornbeck
#47
thisusername said:
Johnson noise doesn't apply because of the function of the dc coupling resistor propagating the current source across its resistance to create the input voltage.
Click to expand...

Basic theory (like Johnson noise) always applies. But theoretical limits can far exceed some practical limits.
For example, we can hear down to around 0dB spl, but you'll have trouble finding anywhere nearly that quiet.
 
  • Like
Reactions: kevinkr
#52
Phono cartridges are voltage sources with a defined (winding specific) series resistance and rising inductive reactance (XL) with frequency. Since inductance is typically mH or less the impedance variation in band is not very significant. relative to the load resistance of 47K. (At 20kHz a 1mH inductor has an XL of approximately 125 ohms so it's typically much less than the DCR of the cartridge or the load resistance)

The cartridge DCR contributes its own Johnson noise to the equation as well.
 
  • Like
Reactions: rayma
#53
rayma said:
I suppose it's just a resistor that passes signal by connecting from one node to another node.
Like a coupling capacitor. This term may be most often used for guitar amplifiers these days.
Click to expand...
When its in between stages it is the signal current load. When the device is a passive element, the signal current applied across the resistance renders a voltage source at that impedance.
 
#56
With the caveat that MM/MI cartridges have much more inductance, maybe half a Henry or so. Fancy MC cartridges, even the high output ones, are certainly comparatively low inductance.

All good fortune,
Chris
thisusername said:
Who really cares about Johnson noise. Its not going to be a main part of the signal. If you are so worried about it get low self noise resistors.
Click to expand...
I think the point being missed is that "low self noise resistors" do not evade Johnson/Nyquist/thermal noise, which is fundamental. They can minimize excess noises, which are related to current passing through the resistor, but even a "perfect" resistor with no excess noise will still have RT noise.
 
  • Like
Reactions: kevinkr and rayma
#57
Hi Chris,
I looked up a typical Grado and found the cartridge inductance of 45mH which equates to an XL of about 5.6K which is getting significant. I am used to working with LOMC cartridges with DCR of a few ohms and inductance in the low uH range, and commensurately low output levels. (200uV - 500uV @ 1kHz @ 3.54cm/sec lateral recorded velocity).

Dubadub - sorry for all of the OT in your thread, please let us know how your noise mitigation goes and whether you need more help.
 
  • Like
Reactions: rayma
#58
Yeah, Grado's have an unusually low inductance compared to typical Shure's and Stantons and such, old skool cartridges. My Shure "Ultra 500", a version of type V, runs about 360mH IIRC. For these guys, the 47K loading resistor becomes a significant noise source because the source inductance doesn't shunt it very well at higher frequencies. There've been several different interesting alternative approaches, but outside of this thread's intent.

Much thanks, as always,
Chris
 
  • Like
Reactions: kevinkr
#59
Thanks to all for the spirited discussion. Nothing better than getting experienced folks together to debate an unsettled topic.

The device in question is a 834p clone built on the Douk board according to the LHTR specs. It's a tight fit in the enclosure and the input jacks are connected to the PCB with two pieces of .999 silver wire about 1" long.
IMG_20231116_134830976~2.jpg



I'd like a discussion of best practices regarding noise in the wiring of the inputs of a phono pre to fight noise. I live in an urban area with gobs of all types of RF, and I'd like to know the best ways to avoid, reject, cancel, etc that noise. Grid stoppers are a good way in guitar amps, but I don't want to lose any more signal than absolutely necessary in a hi-fi amp.

Ferrite beads, extra inter-electrode caps and other things that can be done to the input wiring/PCB seem most attractive. Cutting traces and modifying the PCB itself is less desirable, unless the outcome is superior. Same goes for altering the TT.
 
#60
Nothing strikes me as terrible, the board may be working against you to some extent.

I would start by adding at least one 0.1uF cap from your GND buss bar to the chassis. Next I would consider removing the copper bus depending on how the GNDs are arranged on the board.

Finally I would add a ferrite on the signal lead between the jack and board.

Nothing looks terrible BTW or particularly unreasonable so I would also look at better interconnects to your TT.

You wouldn't happen to have a photograph or two of the pcb that shows how the traces are routed.

You can also add up to something like 220 ohm resistors right at the input grids - 1206 SMD resistors are ideal, just cut the trace very near the grid pin, scrape, tin and them solder the resistor in place. SMD resistors are ideal as they have low parasitics compared to leaded parts. (Panasonic, Beyschlag/Vishay, or Yageo thin films are what I would recommend.)

I would not connect the chassis directly to the PCB in most cases, but I don't know enough about your build to know what I would recommend. (PSU in the same box, safety earth?)
 
You must log in or register to reply here.
Top Bottom