Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

High Line voltage, Old radios

1 view
Skip to first unread message

NS8S

unread,
Nov 7, 2009, 10:27:51 AM11/7/09
to
I'm wondering what the best way to lower the filament voltage of a
simple, 4 tube AM series filament string, line powered would be when
my line voltage funs 122-123 volts, and my 25 volt filaments are
sitting at about 28 volts? Ballast Tube runs extremely hot, most of
them do. Series resistor, high wattage, near power supply? I can get
appropriate voltages with variac, no problem. Radio is much happier.
Thanks for the input. Just looking for a few suggestions.

philo

unread,
Nov 7, 2009, 10:40:04 AM11/7/09
to


A series choke is a good way to go...
the 12v secondary of most any transformer will work as one...
just as long as you do not exceed it's current rating.

Pete Bertini

unread,
Nov 7, 2009, 11:02:56 AM11/7/09
to

"NS8S" <david....@verizon.net> wrote in message
news:28d4c67b-2bf6-48fe...@t2g2000yqn.googlegroups.com...


I'd drop the ballast tube and go with a series capacitor to reduce the
voltage.
The ballast will run hot, regardless of a few volts additional voltage drop.
But, the series cap will eliminate the ballast if you desire. A cap has much
less ESR losses than a choke, either way you will need to calulate the
reactances at 60 Hz and go from there to find the correct value. Use a
good mylar, NOT a cheap bipolar electrolytic AC cap if you do this.

or.,...

You could add a filament transformer wired as an autotransformer to
reduce the AC line voltage to the entire radio by 6 volts. Not worth the
effort.

or....

A series diode on the filament string will drop the voltage by .707 times,
or to about 84 volts for typical US line voltages. You can use a diode
to drop the AC line to about 84 volts, and use a capacitor for the final
reduction to 25 volts from there, the diode generates very little heat and
a good capacitor will also not generate heat.

or.....

If it was my set, I'd consider adding a 24 VAC filament transformer and
dumping the ballast completely if the transformer could be hid inside the
case.

Peter


Bill M

unread,
Nov 7, 2009, 11:06:12 AM11/7/09
to

I guess you probably don't have room for a bucking transformer? I often
use one of those chassis mount resistors. Around 22 ohms knocks off
8-10 volts on a 5-tuber. I got a bunch of them for next to nothing so
thats what I use most often.

Mount it where it won't cook other components! On top of chassis in
open air is best.

nesesu

unread,
Nov 7, 2009, 11:18:04 AM11/7/09
to
On Nov 7, 8:02 am, "Pete Bertini" <radioconnect...@cox.net> wrote:
> "NS8S" <david.flem...@verizon.net> wrote in message

A small point, Peter. You cannot use a diode AND a capacitor at the
same time to drop the voltage to a filament string. You can use one or
the other just fine. To trim the drop using a diode to give somewhat
less than the 84 or so volts you need to used a resistor, and thus the
power dissipated will slowly increase. The capacitor method is,
indeed, the most efficient method and [the current] can be easily
trimmed to provide any desired voltage across the string.

Neil S.

Pete Bertini

unread,
Nov 7, 2009, 11:22:45 AM11/7/09
to

"nesesu" <neil_su...@telus.net> wrote in message

A small point, Peter. You cannot use a diode AND a capacitor at the
same time to drop the voltage to a filament string. You can use one or
the other just fine. To trim the drop using a diode to give somewhat
less than the 84 or so volts you need to used a resistor, and thus the
power dissipated will slowly increase. The capacitor method is,
indeed, the most efficient method and [the current] can be easily
trimmed to provide any desired voltage across the string.

Neil S.

Never thought of that, since I never tried it.. But, what you
are saying makes perfect sense. The cap would block the
DC component I suspect?

Pete


WW

unread,
Nov 7, 2009, 11:26:11 AM11/7/09
to

"NS8S" <david....@verizon.net> wrote in message
news:28d4c67b-2bf6-48fe...@t2g2000yqn.googlegroups.com...

Replace the ballast tube with another radio tube that would have the same
current rating as the other tubes and of voltage needed to cover the voltage
needed.
Just rewire the ballast tube sockt as needed. This should produce less heat.
Warren


Bill Noble

unread,
Nov 7, 2009, 12:15:28 PM11/7/09
to

"WW" <cc...@nospambresnan.net> wrote in message
news:44idnZ5uI-4KAmjX...@bresnan.com...

Warren - I don't see how it is possible that a different tube produces less
heat. Whatever you put in there - a tube, a resistor, etc, will have the
same current, and power dissipation is I*2R - so any resistive element with
the same current and resistance will dissipate the same power.

No one has mentioned the possibility of an active regulator - that would be
the most aesthetic - you could have "soft start" to prevent the huge current
surge into cold filaments, and not worry about line voltage. The power
dissipation would be low if you used a chopper circuit. There are plenty
of single chip, or chip plus a pass transistor circuits out there, and they
are small enough that you can enclose in a piece of copper tubing and solder
shut for good shielding - your typical AA5 radio draws somewhere between 1/4
and 1 amp -

http://www.nxp.com/acrobat_download/applicationnotes/AN120.pdf is a good
overview of such supplies in general.

But for this application you have another alternative - you don't really
need regulation, just a voltage drop, and ideally a soft start - consider a
diode bridge and a transistor as the pass element for the circuit -the
bridge is needed to make sure power always goes the same way through the
transistor - and 1 amp at 200V NPN is a common transistor. When transistor
is on, AC flows through the diode bridge, through transistor, and on to the
radio (you will need to make sure that the output polarity is such that the
DC will pass through the rectifier tube). Now, drive the transistor with a
555 configured as a 50% to 100% square wave, and power it off a
resistor/capacitor from the mains. Drive the duty cycle control through a
voltage divider (adjustable, of course), with at least 50K total resistance,
and include a 100 microfarad cap so the voltage on the duty cycle pin climbs
slowly after power is applied. You now have a chopper type voltage
dropping device with soft start. test it on the bench. Use a reasonably
high frequency for switching (but safely within the transistors
capabilities). Add a small series inductor at input and output (to cut RF
hash). when it works, put it inside a copper tube and use feedthrough
caps -

Kenneth Scharf

unread,
Nov 7, 2009, 1:33:52 PM11/7/09
to
I'm guessing you want to drop the line voltage by about 5 volts or so.
How about putting a pair of series connected zener diodes with the
cathodes connected together? The reverse voltage drop across a zener is
the rated voltage, and the forward voltage drop across a zener is about
1 volt. So a pair of series connected 5.1 volt zeners should drop about
6v AC.

Ken

unread,
Nov 7, 2009, 1:02:25 PM11/7/09
to
Or a 6V fil xfmr, using the secondary as a bucking voltage, this would
trim off 6V. Ken

WW

unread,
Nov 7, 2009, 1:43:10 PM11/7/09
to

"Bill Noble" <nob...@nowhere.invalid> wrote in message
news:hd49vf$8fe$1...@news.eternal-september.org...

>
>
> "WW" <cc...@nospambresnan.net> wrote in message
> news:44idnZ5uI-4KAmjX...@bresnan.com...
>>
>> "NS8S" <david....@verizon.net> wrote in message
>> news:28d4c67b-2bf6-48fe...@t2g2000yqn.googlegroups.com...
>>> I'm wondering what the best way to lower the filament voltage of a
>>> simple, 4 tube AM series filament string, line powered would be when
>>> my line voltage funs 122-123 volts, and my 25 volt filaments are
>>> sitting at about 28 volts? Ballast Tube runs extremely hot, most of
>>> them do. Series resistor, high wattage, near power supply? I can get
>>> appropriate voltages with variac, no problem. Radio is much happier.
>>> Thanks for the input. Just looking for a few suggestions.
>>
>> Replace the ballast tube with another radio tube that would have the same
>> current rating as the other tubes and of voltage needed to cover the
>> voltage needed.
>> Just rewire the ballast tube socket as needed. This should produce less
Bill.. I look for simple ways to do things and I thought a tube would run
cooler than ballast tube.
Going on my memory From when I worked in radio repair shop in the 40's and
50's Warren


terry

unread,
Nov 8, 2009, 9:50:30 AM11/8/09
to
On Nov 7, 1:22 pm, "Pete Bertini" <radioconnect...@cox.net> wrote:
> "nesesu" <neil_sutcli...@telus.net> wrote in message

Oh! When I read that I took it to mean a series diode and a capacitor
across the radio part of the circuit?
In other words the radio fed by half wave rectified AC. But that would
mean a rather low voltage?

My first thought too was a voltage 'bucking' transformer. A 12 to 18
volt front door bell transformer would work.

In fact many years ago (1958 etc.) before the electrical system in
this part of Canada was upgraded to to it's current excellence and
high level of service, fixing black and white TV in rural and often
low voltage situations it was necessary to 'bench test' at low
voltages.

I arranged two such transformers (didn't have a variac back then) with
regular changeover light switches so that line voltage at my repair
'shop' (The 'shop' was, first the porch of a house trailer, later a
closet between two bedrooms and later a whole 12 by 20 self built shed/
workshop.) could be reduced in two steps. And if some of those old
RCA, Admiral, Marconi, Motorola etc. monster TVs in wooden boxes would
work at less than 100 volts they'd work anywhere!

One amusing sidelight was, working in small communities the number of
times one would be told that it was the first TV ever seen in that
community! Where I lived there were at least seven such 'First TVs'!
The reference was usually accompanied by a description such as; "Yes,
I remembers the time we brought her home in back of Joe O'Brien's
truck ...... . It was spitting rain so we wrapped it up in a
tarpaulin .....!" I still meet people, now in their 50s and 60s who
will say they recall, as children, the time I came to their house to
fix their TV.

Very different today; e.g. wall mounted flat screen TV with teenager
playing video games by remote control from a couch/sofa/davenport on
other side of the room.

terry

unread,
Nov 8, 2009, 10:03:10 AM11/8/09
to
On Nov 7, 3:43 pm, "WW" <c...@nospambresnan.net> wrote:
> "Bill Noble" <nob...@nowhere.invalid> wrote in message
>
> news:hd49vf$8fe$1...@news.eternal-september.org...
>
>
>
>
>
> > "WW" <c...@nospambresnan.net> wrote in message
> >news:44idnZ5uI-4KAmjX...@bresnan.com...
>
> >> "NS8S" <david.flem...@verizon.net> wrote in message
> > hash).  when it works, put it inside a copper tube and use feed-through

> > caps -
>
> Bill.. I look for simple ways to do things and I thought a tube would run
> cooler than  ballast tube.
> Going on my memory From when I worked in radio repair shop in the 40's and
> 50's  Warren- Hide quoted text -
>
> - Show quoted text -

Just another thought on using 'door-bell' transformers for voltage
bucking!
BTW saw some 16 volt secondary 14 watt transformers in surplus outlet
other day for $4.
That secondary probably good for one amp. Also factory insulation test
prob 2Kv. or so.
And in past have not perceived any problems with the secondary
insulation, when placed in series with line voltage supply to the
radio, breaking down to either the core or the 115 volt primary
winding.
Also btw although door bell/chime transformers not designed AFIK for
continuous service (just the occasional and momentary stab of the the
push button and maybe a very small lamp within the door bell-push) no
problems with overheating when used for several hours at a time.

Jughead1

unread,
Nov 8, 2009, 10:10:04 AM11/8/09
to
A CL90 NTC resistor will drop about four volts and gives a bit of surge
protection which might prolong the life of the tube filaments. That is what
I use in most of my restorations. Install the CL90 away from heat sensitive
parts such as electrolytics and oscillator coils.

"NS8S" <david....@verizon.net> wrote in message
news:28d4c67b-2bf6-48fe...@t2g2000yqn.googlegroups.com...

terry

unread,
Nov 8, 2009, 10:18:23 AM11/8/09
to
On Nov 8, 12:03 pm, terry <tsanf...@nf.sympatico.ca> wrote:

BTW that 'voltage bucking' idea can be used elswhere. e.g. late one
night/early morning to fix our now 48 year old dryer. Our then teenage
son and I used such an arrangenent to operate a $1.00 but only 78 volt
relay to replace the broken start contacts of the electric motor. It
worked fine for some 15 to 20 years until one of the stator windings
of the motor went open.

Engineer

unread,
Nov 8, 2009, 3:33:22 PM11/8/09
to

All my vintage radios and tube amplifiers are run from fixed ratio
bucked supplies between about 110 and 115 VAC (depending on time of
day/night.) I use no less than four of these for several vintage
radios around the house; two made from 6.3-0-6.3 volt filament
transformers, one from an AC wall-wart and one from a "rescued"
industrial power transformer that happens to have useful low voltage
taps. Work-bench has a proper 1:1 isolated and bucked AC supply
controlled by a variac.
Cheers,
Roger

philo

unread,
Nov 8, 2009, 5:29:18 PM11/8/09
to

I am lucky in that I have a job where I often get to take home
de-commissioned commercial UPS units...

but for the situation the OP is in a simple series choke should do the job.

I'd not modify the radio in any way...
just put the choke (or xfmr) it in a small enclosure

NS8S

unread,
Nov 10, 2009, 9:28:38 AM11/10/09
to

That's great, for me, Roger, however this set belongs to a person that
only wants to play the radio, and has no electrical sense whatsoever.
I'm going to install dropping resistors in the AC line. I've done this
before, clunky, but it works well.
Cheers,
Dave

John Goller, k9uwa

unread,
Nov 10, 2009, 7:27:34 PM11/10/09
to
In article
<201283e3-83c1-4ff8...@v25g2000yqk.googlegroups.com>,
david....@verizon.net says...

>
>
>That's great, for me, Roger, however this set belongs to a person that
>only wants to play the radio, and has no electrical sense whatsoever.
>I'm going to install dropping resistors in the AC line. I've done this
>before, clunky, but it works well.
>Cheers,
>Dave

Probably 75% of the radios that leave my bench have some sort of line drop
arrangement in them. All are expected to be played from regular AC outlets.
Most get resistors in the AC line and higher tube count radios will get a
bucking transformer. Ballpark we do about 75 radios per year with close to
half of them being back to their Non Tech Owners.

John k9uwa

0 new messages