[Amps] Henry 2k-4 HV inductor - Tuned choke input filter

Jim Thomson jim.thom at telus.net
Fri Dec 15 13:12:23 EST 2017

Date: Fri, 15 Dec 2017 01:33:24 -0700
From: John Lyles <jtml at losalamos.com>
To: amps at contesting.com
Subject: [Amps] Henry 2k-4 HV inductor - Tuned choke input filter

<At the time (1980s) no one would have put a huge capacitor filter in the 
transmitter, due to cost and space. Voltage for a single phase 3500 watt 
transmitter was 4500 VDC. Series electrolytics did not meet the long 
term 24/7 reliability goals. 

<One other point, I learned that the resonant choke doesn't provide 
adequate filtering for harmonics of the rectification frequency. 240, 
360, 480, etc. They just roar through the L as it has a shunt C 
bypassing it for the harmonics. So a second L-C section had to be added 
to the design, for the higher order filtering. It was very small, not 10 
Hy or anything like that.

<All in all, it was an interesting project, but I would never do it again 
today, well who would use a tube for FM anymore? It is probably not an 
appropriate design for SSB or AM where the load is varying significantly 
all the time.


##  The ripple current ratings on these modern day electrolytics is
a quantum improvement from years gone by.  Per the various tech notes from
hammond, sprague, CDE, united chemi-con, etc, the consensus is the CCS
  DC load current should be no more than the... CCS ripple current rating divided
by 2.56 

##  On paper, a string of lytics with a 10A   CCS ripple current rating should be good for
a max of  3.90 A   CCS plate current.   These days, you can get ripple current ratings a lot
higher than 10A, in a high C,  450 or 500 vdc lytic.  You can also parallel a 2nd  or 3rd string
for more CCS ripple current.   For something like a FM broadcast TX,  running 24 hrs, 365 days
per year,  you would want it ultra conservative.   The tech notes depict formulae for determining 
how many years to expect from modern day electrolytics.   CCS current drawn, ambient air temp
surrounding the circumference of the lytics, loaded  B+..as a percentage of the max 450 vdc rating,
EQ resistor value, internal leakage current, cap value in UF, etc, etc, is about all you require to come up with a good 
estimation of electrolytic life.   If you make sure the ambient temp inside the  HV cabinet is low, like 
the same as the ambient room temp, and also run the lytics at only 70-75 % of their max stamped V rating,
and dont max out the ripple current, they will last a long time, like > 15 years.  

##  They showed examples for stuff like freq drives and   old style telco type  52 vdc rectifiers, where 
365 operation was planned, and min acceptable service life used for the tech example was 9 years. 
And how to re-calculate for say 5 years, or  15 or 20 years.   Heat played a big factor..which is easy to 
remove  from a HV cab. 

##  I remember you mentioned that the resonant choke would not filter out the harmonics of the  rectification
freq...and that additional LC sections were required.  It was to do with the.....  AM noise spec on a FM broadcast TX. 
The 240-360-480- 600 etc, etc, had to be reduced to an acceptable min level. 

##  A simple high C filter will kill the fundamental..and also all the harmonics.    The last B+  supply I had used 4 
CDE oil caps in series parallel.    Each was  stamped 47.6 uf  at 4.5 kvdc..and each weighed 30 lbs,so 120 lbs for the 4 x oil caps. 
So ended up with 47 uf at 9 kv.  The 4.8 kv sec of the 10 kva hypersil pole pig and FWB, netted +6700 vdc  no load.  
We paralleled an additional 100 uf via a string of  20 x 2000 uf lytics.   Its now being used on a hb  80-10m  YC-156.  

##  But you are correct, any new FM broadcast TX in the last 2 decades uses SS, typ in 1 kw modules,  complete with mating 
50 vdc switching supplies.   Those  -52 vdc switchers  we installed at the telco, just b4 I retired were rated for 65A  CCS each,
didnt even run luke warm, and  the 208 vac single phase versions  we used had a > 99.9 Power factor at any phase angle. 
They would alos run just fine on anything from 140 vac up to 300 vac..and any freq from 40-70 hz.  Typ   6 x of these 3.3 kw
supplies were installed  beside each other in a  25 inch wide rack.   Then row upon row of them, to fill up an entire bay. 

##  Dirt cheap at $475.00 at the time..and  6 x units per shelf were good for 20 kw CCS.  Only 6.5 inchs tall.   The 50 vdc
supplies I see for typ FM broadcast TX are very similar.   And if 1 craps out, the remaining 5 take up the slack.  And in all cases,
the current divides up equally.   They can also be pulled hot.   Old style  20 kw rectifiers were huge things, like 5 foot tall, 
24 inches across, and rated for 50 vdc at 400 A  CCS.   And here a tiny shelf of switchers will easily replace the old style behemoths. 

##  Some outfit in Halifax, in eastern Canada makes a 53 kw  FM TX, consisting  of  55 x 1 kw SS modules and  a 
myriad of combiners. They can be configured from anything from 1 kw to 11 kw in a bay, and multiple bays  for 
a high powered unit.   One RF module craps out, the remaining RF decks take up the slack, and po divides equally. 
RF modules and mating switchers can also be pulled hot.
Tubes are history for FM broadcast, except for legacy equipment.   AM broadcast is barely alive in Canada, they
all moved to the FM band years ago.   We only have 1  AM station left here on vancouver island...well established
talk radio format. 

later... Jim  VE7RF   

More information about the Amps mailing list