[time-nuts] Lead acid battery noise levels

Poul-Henning Kamp phk at phk.freebsd.dk
Tue Jul 9 02:31:01 EDT 2013


In message <0EF755EF-06B6-4543-B0A5-24517D471333 at yahoo.ca>, Mark Spencer writes
:
>Hello, I've just reviewed the thread from earlier this year regarding
>low noise power supplies and some related information about the
>noise levels of certain batteries.    Has anyone ever come across
>any data pertaining to the noise levels of common gelled lead acid
>storage batteries ?

I did some research too, including asking a bloke who does lead-acid
for nuclear reactors for a living.

If you want your batteries to last, the elctrolyte has to move, otherwise
you get stratification with higher density at the bottom, lower density
at the top, and the plates over- and under-charged at the same time.

The noise you get is pretty LF, but it can be quite a few mV and
it is asymmetric, in that positive slopes are generally steeper
than negative slopes.

Gelled and "glass-mat" batteries have higher noise-amplitude but
lower frequency than liquid lead-acid batteries, because the
mechanical resistance to electrolyte movement is much higher, so
it tends to happen in "burps" whereas a liquid lead-acid allows the
electrolyte to flow (almost) freely.

So the recommended configuraiton for low-noise is a liquid lead-acid
("OPzS") which is float-charged in the high end of the voltage range
recommended by the manufacturer.  This gives most electrolysis and
therefore most electrolyte movement and need for water replenishment.

It also means gaseous hydrogen, which requires ventilation.

I havn't done any experiments to verify this.

>I did find some comments on an audiophile forum that indicated
>that gel cell batteries are not known for low noise levels (:

Don't even think about advice from audiohomoeøpathicists, the 99.9%
which isn't downright bullshit is incorrect.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.


More information about the time-nuts mailing list