Amps
[Top] [All Lists]

Re: [Amps] resonant choke filters.

To: 'Jim Thomson' <Jim.thom@telus.net>, amps@contesting.com
Subject: Re: [Amps] resonant choke filters.
From: rejean leveille <va2am@videotron.ca>
Date: Sun, 21 Feb 2010 21:54:49 -0500
List-post: <amps@contesting.com">mailto:amps@contesting.com>
Hi Jim

I like to read expert people on amplifier and I am on a learning curve.

I own since 4 months a Marconi BH 30 modified for the 9 amateur bands,
updated QSK, positive and negative keying, opto transistor links the bias
circuit for a fast response time, harmonic reduction by adding low-VHF Q
parasitic for 10-12 meters, modified bias, in-rush current limiter on 4 of 5
transformers and more.

About Hv filtering, transformer working inside its 80% current limit will
improve IMD from lower hv fluctuation.
My hv transformer is 3 kva 1A going to a full bridge sealed block, 8 mfd/4kv
oiled cap, for regulation and filtering 2x 50k 100w in series with 100 ohms
7w to 2x diodes (1N5408) reversed (double conductivity) in parallel to the
ground. Just before the diode going to the bias cut off circuit, a 9 uh coil
choke connects the other side of the bridge the negative side oiled cap.
Also with this +hv output there is in serie a 10 ohms 10w as current limiter
going the tubes.

There is a variable speed muffin fan to escape the heat to the rear,
monitored with many TC these transfo, HV bridge, tubes, safety switches
would kill from 24v dc both coiled power and hv breakers. Standby mode means
no QSK activation, no hv output, no 110v ac supply, 5v is on for the tube
filaments.

Hv, like other transformers are from Hammond co, multi input taps 100-260
vac, output is 2500 & 3000 v selection.
BH 30 uses 2x 3-500z, plate voltage gives 3450v dc, key down with 65w
excitation, 710 ip, it only drops 225v for easy +1.5 kw output 80-10 meters.
More drive is required for 160, 15 and 12 meters. Tubes are estimated 15
years old.
I am well inside grid current, simple hv circuit, rf deck is built for +3
kw, HD 1/2" size waifer contact on a 4" ring to select the band coil.
The PI-network circuit uses a steel bar plate coil, metal strap Teflon
coated, vacuum variable cap 500-5kv, door knobs cap are 3x in parallel to a
cooper disk to split the current and preventing any capacitance drift done
by temperature. All transformers got many varnish layers but for the
filament it is sealed in epoxy, very thick layer that gives a golden color
and I never saw that before!

The tubes get cooled from a variable speed squirrel fan cage, air filter,
air dock angle with smooth curve to a Y these 2x Eimac glass chimneys under
equilibrate pressure. Fan motor is insulated with rubber gasket to keep low
the noise.
Specs are -20 to +50c, 500w AM, 1000w RTTY 100% duty and 1500w RTTY-SSB 75%
duty for the full temperature range, -46db IMD harmonics, it weights 160 kg
(350 lbs) on wheels and it is very easy to split the rf deck from the power
box for servicing. Original servo motors auto tuning were removed and home
made band selector switches in serie, LED for tx and all 9 bands. 10:1
vernier vacuum condenser, light bulbs for the meters added. Very difficult
to see this is a +30 years old transmitter. It was fully cleaned in and out,
external panels painted and waxed, no rust trace at all.

Entering a contest in cw or RTTY as running mode at 1.2kw or 1.5Kw SSB for
hours, it just get warm on the top cover and stay cold elsewhere. You just
forget it in the corner!

The theory is much simple to apply by using over rated components.

Regards, 73, Rejean VA2AM

-----Message d'origine-----
De : amps-bounces@contesting.com [mailto:amps-bounces@contesting.com] De la
part de Jim Thomson
Envoyé : February 21, 2010 7:08 AM
À : amps@contesting.com
Objet : [Amps] resonant choke filters.

Date: Sat, 20 Feb 2010 11:29:34 -0600
From: "Roy" <royanjoy@ncn.net>
Subject: Re: [Amps] 120 Hz hum from Henry...


"Be careful if you fix this. It's not tuned to *exactly* 120 Hz because that

would cause a very high voltage to appear across the choke. Use the
specified values if 
you replace anything."


I strongly believe the above is a perpetuated old hams' tale, totally 
untrue. The parallel-tuned circuit is a frequency-rejection filter, 
exhibiting a very high Z to the 120 Hz ripple, thus preventing high 
circulating current and its resultant voltage build up. Read the article, 
noting the careful tuning to exactly the 120 Hz resonant frequency: 
http://www.qsl.net/i0jx/supply.html

73,   Roy    K6XK 

&&& IF you resonate the filter to exactly 120 hz... the resonating  cap
WILL  blow up in your face. Try it and you will see for yourself. Been
there, done that.    Ask Tom Rauch  abt his resonant choke experiences,
he just abt killed himself. 

&&&  also note in the I0JX article.  He uses SIMULATIONS. That`s
fine, but apparently the simulations  don`t show the exploding cap. 

&&  also note he advocates using swinging chokes. Swinging chokes
are a dead loss, if used in resonant filters. Their inductance is all over
the map,
exactly what you don`t want.  Dahl`s chokes, made for resonating, are very
stiff, and
 only drop down in value a small amount, when a huge load is applied.   

&&  Look at his plane jane  C input filter.. just ONE 15 uf cap.. 6450 vdc
No load,
and  6175vdc, sucking 1000ma.    6147 divided by 6450 = .957   That`s  a
 4.3% drop. That`s  superb regulation.   If he had increased the C  to
50-150uf,  you
 will then get a real eye opener. Loaded V is higher.. and on SSB, it won`t
droop at all.

In his henry 4 K ultra  example... his No load = 4525vdc.  His loaded V =
3950vdc. 
3950 divided by 4525 = .873    This is a 13%  DROP !    So much for the
nonsense abt
the infamous Henry radio resonant chokes.They are heavy alright, but they
are not heavy duty.

The 8 K ultra used a resonant choke.  It also sucks 300 w of bleeder, just
to make the choke,
look like a choke. ( 5 x 20 k, 100 watt  WW`s in series)  No load is 5400
vdc.  To get that,
they use a 6000 vac xfmr.   If that had been a simple C filer, only a 3800
vac  xfmr is needed...
(and NO choke, nor 300 W  CCS  bleeder)   Loaded V on the 8k = 4800 vdc ,
4800 divided by 5400  = .888    again, no fantastic regulation here. (11.1%
drop) 

IMO, ur better off to dump the resonant choke,and spend the money on the
xfmr, and caps,
not a bunch of excess weight.   Besides, any resonant choke set up will not
kill the harmonics
very well.   The plane jane... BIG C input filter, will kill the 120 hz
ripple,,, and ALL the
 harmonix of 120 hz.  

later... Jim  VE7RF
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps


_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps

<Prev in Thread] Current Thread [Next in Thread>