PT to Choke - Crazy Question .....

[kagliostro]

Sometime I read about people that use (to contain spending) low voltage PT as choke connecting only the primary, although this is far from being an optimal solution

I was thinking to the fact that a PT has a primary and secondary and wondering if something like to put the primaries and secondaries in series will help to improve the situation when a strange idea come to my mind .... the use of a non polarized e-cap (an e-cap because it can be large) in parallel with the secondary



In my mind this capacitor will store a charge and react to flutuations on the primary ........

Is this thing totally crazy ??

Grazie

Franco

[silverfox]

just looking at the circuit posted it seems to me it would oscillate, gently. I too read some articles, just now, and what came up is- Yes, it can be done and the secondary load on the PT in question should be adjusted. I would think a resistor would be a better choice in the secondary though. This is new to me so I look forward to other replies.

silverfox.

[PRR]

You are trying to tune the L-C to 100Hz to improve ripple reduction.

And it will increase higher ripple harmonics.

Since both L and C are low precision parts, it won't stay tuned with current or age.

Put the capacitor money on the end of the coil. Simple and it works.

[kagliostro]

Thanks PRR

Franco

[PRR]

But see also:

http://www.diyaudio.com/forums/tubes-valves/302323-6fm7-pp-amp.html
http://www.diyaudio.com/forums/tubes-valves/302323-6fm7-pp-amp.html#post4953804

[kagliostro]

Thanks PRR

that is really interesting

The 2.2 uF is a "trick" that is usually overlooked, and misunderstood.

This is a special case of a capacitor input power supply (this supply has a 220uF cap input).

A 2.2uF cap will resonate with 0.8 Henrys at 120 Hz.
A 2.2 uF cap will resonate with 1.15 Henrys at 100Hz.

120Hz or 100Hz is the ripple frequency of the full wave rectifier in this circuit.

A parallel circuit of 2.2uF and 0.8H or 1.15H will be a very high impedance to the ripple frequency. It will attenuate the ripple by a very large amount.

Without the 2.2uf, it would require a much higher inductance to get the same reduction of ripple.

A 5H, 10H, or 20H choke is more expensive, larger, and heavier than a 0.8 or 1.15H choke.

Another clue about this supply is that the 2.2uF is not an electrolytic cap.

But do Not try this trick with a Choke Input power supply configuration.
It may cause shorted diodes, shorted cap, or shorted choke turns.
The voltage rise in a Resonant Choke Input supply is Very large.

Franco

[tubeswell]

Very cool.


Thanks for the link PRR

[silverfox]

I'm trying to remember what the exact details were- In the 70's I worked for a company that manufactured Intercoms for hospital use, American Zettler. Anyway, one of the aspects of the power supply was a gigantic power transformer with, as I recall, an AC capacitor across one of the secondary windings; The purpose of which was to prevent a short circuit from destroying the power transformer. I never understood the application but the term used to describe it was: Ferro-resonant. Is that any relation to the above posted schematic with a cap in parallel with the choke?

silverfox.

[kagliostro]

Ciao Silverfox

The idea I had was based on a very old thread (that I didn't remembered well when the idea come out)

some time ago I recovered from a Jerrold 900-B Sweep Generator a Constant Voltage Transformer (read what is write on the transformer)



and at that time I asked what it exactly was here on the forum

(it was a PT with a dedicated winding to which was connected a capacitor)

here the link where you can read an explanation

http://el34world.com/Forum/index.php?topic=12025.msg111313#msg111313


Franco

[Retrovert]

I posted this over at diyAudio in response to that "Super Sub" schematic, and think it is appropriate to make sure people read this before building this power supply.

A detailed article on resonant choke supplies may be found here: www.qsl.net/i0jx/supply.html

One might well ask why every supply doesn't use a resonant choke. The reason is that a resonant choke only filters a specific frequency, typically 120 Hz or 100 Hz depending upon the mains frequency (2x base frequency).

Because the notch is narrow a resonant choke readily passes other frequencies and adds higher-order harmonics, albeit at lower magnitude. Since most of the noise is twice the mains frequency, the resonant choke is reportedly a good way to remove it as long as additional filtering is performed to remove the ripple at the non-resonant frequencies.

Filtering also depends upon load.

As a result, the power supply above described does not properly filter other harmonics and ripple current. It consequently is far less clean than a non-resonant supply, and does not work the way one might think.

The resonant frequency (f0) is given by:
f0 = 1 / (2 × Pi × sqrt(L × C))
Given L and f0, C is determined as:
C = 1 / (2 × Pi × L × f0)^2
Note: L in Henries, C in Farads, and f0 is in Hz.

You may easily determine how permeable the capacitor or choke (inductor) are at any given frequency:
Inductive Reactance: XL = 2 × Pi × f × L
Capacitive Reactance: XC = 1 / (2 × Pi × f × C)
Note: L in Henries, C in Farads, and f is in Hz.

Resonant choke supplies can be very dangerous.

The high voltage across the inductor has 120 Hz or higher order components and it may be boosted to several times the expected value. This can blow up the capacitor if it is not properly rated, can fry the power supply, the amplifier, or you.

This voltage is not readily discharged when the power supply is turned off. This is why a bleeder resistor is required.

Tinkering with high-voltage supplies when one does not understand the ramifications can result in injury, disfigurement, or death.