[AMRadio] How to ID a swinging choke?


jeremy-ca km1h at jeremy.mv.com
Sat Dec 8 16:54:11 EST 2007


----- Original Message ----- 
From: "John Coleman" <jc at pctechref.com>
To: "'Discussion of AM Radio in the Amateur Service'" 
<amradio at mailman.qth.net>
Sent: Saturday, December 08, 2007 2:51 PM
Subject: RE: [AMRadio] How to ID a swinging choke?


Carl,
Saturable reactors "SWINGING CHOKES" and saturable transformers are
strange in the way they operate.  A choke does not exactly lose it reactance
as the DC current is increased thru it but instead it loses its ability to
react over a certain period of AC cycle and AC current level as the DC
current is increased.  The result in XFMRs is similar to clipping of lower
frequency AC/audio.



John, the theory is explained in detail in a series of articles in 1932 
QST's. It even includes an elaborate test setup with AC and DC voltage and 
current meters ( 7 in all) and discussions on setting the gap for different 
loads.

At this point all I wanted was a way to tell the two types apart and a very 
simple test setup.

Im afraid it wont be that easy although the reply earlier and yours gives me 
a few ideas.



In a swinging choke input power supply, the result is to pass large amounts
of current to charge the filter capacitor during the saturation period.
This saturation period is only a part of the period of the input ripple of
the power supply, perhaps only the top 10 degrees at the peak.  But as the
DC current drain becomes higher by demand from the load then the core is
saturated for a longer period of time.  A greater amount of the peak in the
ripple is passed unimpeded to the filter capacitor causing it to charge to a
higher voltage in hopes to compensate for the resistance losses in other
components.  The change in reactance is apparent change over short periods
of the input cycle ripple.  It is actually possible, with certain component
choices, for the output voltage to rise higher than the average voltage at
peak current.  For this reason the output capacitor of the power supply
needs to have a long time constant to absorb slow rises and falls in voltage
as discussed in previous messages.



The input swinging choke critical inductance is the first step that needs to 
be calculated. Then an "optimum" value which is about twice the critical L 
is the value of the choke needed. Optimum and critical L is directly 
proportional to the load resistance and is independent of voltage and C1 
capacity. At this point the AC current is close to the minimum and the 
rectifiers can deliver about 90% of their peak ratings. This is an important 
criteria when using MV tubes in particular. At the same time regulation is 
at its best. The first capacitor (C1) value is not important as long as it 
isnt resonant at 120Hz with the choke;  4-6 mF is more than sufficient for a 
modulator supply and final ripple filtering will be the job of the smoothing 
choke (L2) and the second capacitor (C2).

My particular problem is that the L1 I had known to be a 5-25H turned out to 
be bad. Now I have several unmarked and even sealed chokes to chose from, a 
couple of which are in the 25-30 H range and of sufficient heft to be 
useable for a pair of 810's.

I suppose I can set up a dual trace scope and run a low voltage unfiltered 
DC as well as an adjustable load duplicating the Class B stage. Then see 
what the AC and DC waveforms do.

Final step will be L2/C2 and getting the ripple down below any possibility 
of getting on the carrier. I understand that PP modulators are supposed to 
cancel out small amounts of ripple but Ive no experience in that area.

Dont mind me, Im trying to think my way thru this as Im typing. Ive found it 
sometimes works!




So, in short, it is difficult to measure the effects of saturation with out
a scope, a variable AC and variable DC supply, as well as having some known
examples to work with as well.


I believe I have everything I'll need to set up the test.
Thanks for letting me ventilate!

Carl
KM1H




Don, K4KYV, had come up with the idea of taking the "I" section of iron from
two identical 12 Henry chokes.  He did this and placed the two "E" section
together, series connected the two windings in the proper phase.  The idea
was to make modulation reactor of about 40 Henry.  The gap between the two
sections had to be adjusted for the maximum reactance at the DC current
level that final amp was using.  We did this by series connecting a known
choke with the experimental choke and actually using the two of them as a
modulation reactor.  Using a insulated VOM to measure the AC voltage drop
across each and adjusting the gap we were able to come up with a very nice
modulation reactor but is was tedious and dangerous working with the HV
stuff and making sure the DC was all gone before adjusting the gap or
changing the meter probe points.  Thanks still to Don as I am still using
it.

I am sure some type of test facility could be constructed for this with some
imagination

John Coleman, WA5BXO

-----Original Message-----
From: amradio-bounces at mailman.qth.net
[mailto:amradio-bounces at mailman.qth.net] On Behalf Of jeremy-ca
Sent: Saturday, December 08, 2007 8:16 AM
To: amradio at mailman.qth.net
Subject: [AMRadio] How to ID a swinging choke?

Is there a simple bench test or an eyeball way to tell the difference
between a fixed and swinging choke? Something about an air gap if I remember

but not all my chokes are open frames.

Carl
KM1H

______________________________________________________________
Our Main Website: http://www.amfone.net
AMRadio mailing list
List Rules (must read!): http://w5ami.net/amradiofaq.html
List Home: http://mailman.qth.net/mailman/listinfo/amradio
Help: http://mailman.qth.net/mmfaq.html
Post: mailto:AMRadio at mailman.qth.net
To unsubscribe, send an email to amradio-request at mailman.qth.net with
the word unsubscribe in the message body.

______________________________________________________________
Our Main Website: http://www.amfone.net
AMRadio mailing list
List Rules (must read!): http://w5ami.net/amradiofaq.html
List Home: http://mailman.qth.net/mailman/listinfo/amradio
Help: http://mailman.qth.net/mmfaq.html
Post: mailto:AMRadio at mailman.qth.net
To unsubscribe, send an email to amradio-request at mailman.qth.net with
the word unsubscribe in the message body.



More information about the AMRadio mailing list

This page last updated 19 Oct 2017.