From borowsk–(at)–pk.hp.com Thu Sep 14 08:30:39 CDT 1995
From: borowsk–(at)–pk.hp.com (Don Borowski)
Subject: Re: Caps, caps and more caps.
Sender: new–(at)–pcvsnz.cv.hp.com (News )
Date: Wed, 13 Sep 1995 19:21:21 GMT
Xref: geraldo.cc.utexas.edu rec.audio.tubes:1271 alt.guitar.amps:3621
Reid Kneeland (rei–(at)–oldev.tti.com) wrote:
: Well, as long as we’re asking stupid cap questions…
: Are mica and silver mica caps the same thing (in other words,
: are all “mica” caps really “sliver mica” caps)?
: Along the same lines, are polypropelene and metalized polypropelene
: the same thing, or are there non-metalized ones as well?
Non-metalized capacitors use actual metal foils as the conductors.
Metalized capacitors have metal deposited (usually by evaporation)
onto the dielectric.
Since the metal on metalized capacitors is so thin, the capacitors can
be made smaller. Also, the metalization can blow like a fuse if there
is a localized short, so there is “self healing”.
On the other hand, non-metalized capacitors can handle higher current.
: Third, is there any harm in adding more capacitance to a power
: supply? I’m talking about a factor of 3 or so, like from 50 to
: 150 on the first cap after the transformer. The rectifier in
: qestions is solid state, so the switch-on current shouldn’t matter.
: I know that the benefit is debatable, but is there any actual harm?
: (I just happen to have the bigger caps lying around, so less capactiance
: would actually cost me more money at this point.)
You have to be careful here. A lot depends on the impedance at the
secondary of the transformer. If you have a high quality, low impedance
transformer, increasing the capacitor value will increase the peak currents
seen by the transformer, rectifier, and capacitor. Since presumably the
first two are unchanged, they will run hotter, and thus be less reliable.
If you can stand to lose a little voltage, you would do well to put some
resistance in series with the rectifier diode(s). Even as little as 10
ohms can do a lot for reliability.
: And finally, how does one go about choosing a smoothing choke
: value to put into a supply that doesn’t currently have one? Is it
: closely related to the voltage and capacitnaces being used? OK,
: that wasn’t a cap question, but it’s related (inversely).
It is really related to the voltage and current. If you are building
a choke input filter, the rule of thumb is to select the choke to be
equal to the output voltage divided by the output current in mA. Thus,
for 300 volts out at 100 mA, you will need a 3 henry choke. If the
choke is made smaller, you will have poor voltage regulation. For an
amplifier, you would need to select the choke for the idle current.
Note that a choke input filter will give you (in theory) a dc output
voltage of .9 times the rms rectified voltage. Contrast this to a capacitor
input filter where the dc output voltage is 1.4 times the rectified rms
When using chokes after a filter capacitor, there is essentially no
critical minimum value as outlined above.
Donald Borowski WA6OMI Hewlett-Packard, Spokane Division
“Angels are able to fly because they take themselves so lightly.”