DIY Audio Projects Forum

hum diagnostic
Page 1 of 1

Author:  Gab [ 16 Apr 2019, 18:15 ]
Post subject:  hum diagnostic

hi everyone.

I have what I hope is a simple question. Answer to these questions are usually simple too :D

I'm having trouble diagnosing a 120z hum problem. Is there any way to know if i'm dealing with power supply ripple or a ground loop ? Is there some test that can be done in order to know if it's one or the other without any doubt ? I'm still dealing with hum problems with my 4S and i've been trying a lot off diffrent things to solve my problem. The hum i'm getting right now is nothing to the one I had when I first tested it, but it's far from perfect and i'd love to bring it to a better level.

Searching for informations about my problem, I realized that there is a lot of information about avoiding hum, but there is practically nothing about finding hum problems for the DIY tube hifi enthusiast. It could be a wonderfull article to go with the power supply and grounding techniques articles ;)

Thanks for your answers, and i'm ready to help if anybody wants to write an article on the subject !

Author:  Suncalc [ 16 Apr 2019, 19:23 ]
Post subject:  Re: hum diagnostic

In general, hum at the mains frequency (60Hz in your case) is due to grounding issues. It means you are picking up a mains frequency sine wave from somewhere. Hum at twice the mains frequency (120Hz in your case) is almost always a filtering issue.

Occasionally a 2*Fm hum can be due to high EM field spikes if the first capacitor is too large (especially when using solid state rectifiers.). This drives the rectifier conduction angle down and can make the secondary currents in the feed lines couple into places they shouldn't; especially if there is too much stuff put in too small a box.

Personally I would look at power supply filtering first. Check for cold solder joints and make sure everything looks like it's wired correctly.

Author:  Gab [ 17 Apr 2019, 09:26 ]
Post subject:  Re: hum diagnostic

Do you think that I could use this filter module with a tube rectifier ? I’ve bought it tondo some tests but i don’t know if it’s ok for a tube rectifier. I could also replace the last resistor with the choke i already have...

If it’s not usable in my application, i’ll just rebuild the whole power supply. I have spares to build a sexond one so there will be no problem with that too.

Author:  Gab [ 17 Apr 2019, 15:32 ]
Post subject:  Re: hum diagnostic

The chinese board definitely gives better results than my power supply as it was before. No more hum. There is what I would call noise, but i’m pretty shure that it comes from the 3$ pot I used. Stepped attenuator is on the way...

By looking at the circuit diagram posted earlier, could somebody tell me if it’s possible to replace R3 by a choke ? Thanks !

Author:  ILoveHiFi [ 17 Apr 2019, 19:05 ]
Post subject:  Re: hum diagnostic

If you add additonal choke filtering and another 22uF would be even lower noise levels.
You can replace r3 by choke but quite meaningless

Author:  Gab [ 17 Apr 2019, 20:17 ]
Post subject:  Re: hum diagnostic

ILoveHiFi wrote:
If you add additonal choke filtering and another 22uF would be even lower noise levels.
You can replace r3 by choke but quite meaningless

After a good listening I doubt that I would benefit from more filtering. I’m curious to build a new PS as it was before, on a perforated boar with new parts and cleaner wiring, just to see what it can do. I also could call it a day and instal the chinese boad permanently since it seems to do its job quite well for the 7$ I paid. (Wich is approximately the price of the nichicon 100uf I used in the last setup...)

Author:  Peter W. [ 19 Apr 2019, 12:18 ]
Post subject:  Re: hum diagnostic

Some basics:

a) 120Hz hum is very nearly always filtration. It is not: Rectifier - that would be 60Hz, It can well be a ground loop - try to star-ground rather than daisy-chain ground, that mostly cures such things.
b) 60Hz hum is very nearly always a bad rectifier. Under rare and unusual conditions, it could be a unique form of induced hum - where a signal cable runs too close to an AC source.
c) Induced noise will manifest as many things, not just hum. And can be indicative of failing caps, transistors going open (or short) and poor connections/cold-solder joints.
d) Try to shield your wiring in logical sections, and, especially, try to shield the power-supply from any signal wiring.

Means and methods include:
1. Twisted pairs - any signal wire should be twisted.
2. Metal shielding - sheet-metal the depth of the chassis as a barrier between sections. Penetrations should be grommeted, and mixed classes should not share the same penetration.
3. Shield wires need not be connected at both ends in many cases - it depends very much on the actual application.
4. Wire path is critical. Sometimes removing hum is as simple as re-routing some wires.

One thing: If you get it right, but note the steps you took to get there, the next time it will be much easier. And the third time, it will be instinctive. So, keep track of those steps!

Author:  Gab [ 20 Apr 2019, 21:10 ]
Post subject:  Re: hum diagnostic

Thanks ! Very instructive !

Page 1 of 1 All times are UTC - 6 hours [ DST ]
Powered by phpBB® Forum Software © phpBB Group