• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Simple AC mains line filters (DIY, Schaffner ...) and some measurements

pma

Major Contributor
Forum Donor
Joined
Feb 23, 2019
Messages
2,372
Likes
4,352
Location
Prague
Following the current discussion on AC mains filter I have briefly completed a simplest LC filter AC mains filter sample.

!!Please do not follow to build the example I am showing!! It is dangerous with lethal voltages present. It looks like this

IMG_1159.jpg

and it is the LC filter, L = 9mH, C = 235 nF

Here is the measured frequency response. Generator output impedance is close to 0 ohm, load impedance is 3 kohm.
AC_line_filter.png


Noise and distortion of SS preamp and tube preamp were measured with and without the filter, similar as done by @amirm . No differences were found with the SS preamplifier.

Some slight differences in output noise spectrum of the tube preamplifier were found, as can be seen below.

tubepre noise nofilter.png


tubepre noise ac mains filter.png


Please feel free to suggest further measurements.
 
OP
pma

pma

Major Contributor
Forum Donor
Joined
Feb 23, 2019
Messages
2,372
Likes
4,352
Location
Prague
I tried more measurements and experiments with the tube preamp and AC mains filter vs. standard AC power cord. Like switching light bulbs, electric oven etc. No notable difference was found at the tube preamp output when the preamp was supplied with/without the filter.

Finally, I made a test with a 1kHz sine tone, with/without the filter, recorded the files and applied Paul's @pkane DeltaWave software on the recorded waveforms. The results are below and indicate no audible difference. The first plot is measured with REW to show the test tone level at tube preamp output and distortion numbers.

Tubepreamp.png



tubepre_filtertest_origspectra.png


tubepre_filtertest_matchedspectra.png


tubepre_filtertest_spectrumofdelta.png


tubepre_filtertest_pkmetrics.png


The difference is within the parameter/noise drift of the tube preamplifier.
 
Last edited:
OP
pma

pma

Major Contributor
Forum Donor
Joined
Feb 23, 2019
Messages
2,372
Likes
4,352
Location
Prague
Effect of generator impedance (0 ohm vs. 50 ohm) to HF mains filter frequency response (Schaffner FN2020-1):

mainsfilter_generatoreffect.png


This is a standard mains filter with current-compensated choke, so it is effective to common mode interference and not very effective to the differential mode interference, when the magnetic fluxes cancel the effective inductance, as can be seen from the measurement and image below.

1630387449493.png


1630388164820.png
 
Last edited:
OP
pma

pma

Major Contributor
Forum Donor
Joined
Feb 23, 2019
Messages
2,372
Likes
4,352
Location
Prague
Common mode and differential mode interference explained here above.

common_mode_diff_mode_response.png


Resonant peak in the differential mode with low Zgen indicates that under certain conditions the filter may worsen the interference noise rather than improve.
 
Last edited:
OP
pma

pma

Major Contributor
Forum Donor
Joined
Feb 23, 2019
Messages
2,372
Likes
4,352
Location
Prague
Now what is the effect of the simple LC filter 9mH+235nF from post #1 to the mains spectrum. Mains spectrum was measured without the filter and with the filter, measuring in differential mode.

1) without the filter
without_filter.png


2) with the filter
with_filter.png


We can see some improvement above 5kHz. We can see a resonant peak near 3.5 kHz, exactly as in post #1 frequency response. We can see no improvement below approx. 2kHz.
The use of such filter for audio is highly questionable, it may even worsen the residual noise due to LC resonance. No improvement was measured at the ouput of SS and tube preamplifiers. No improvement was measured when switching on-off bulbs and domestic appliance on the mains.

The filter used in post #2 (Schaffner FN2020-1-06) was even more problematic. With a CM action starting above 10kHz and DM resonant peak above 30kHz (where we often can see the SMPS operating frequency) it would be useless for audio. These components are designed for different purposes than audio.

In audio the only helpful AC mains components are 1:1 isolation transformers in case we have troubles with power ground loops.
 
Top Bottom