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 Historical content remains viewable.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

TRANSFORMER QUESTION

7 views
Skip to first unread message

fleis...@comcast.net

unread,
Nov 23, 2009, 10:47:03 PM11/23/09
to
Hi all,

How warm should a transformer in a 40's console radio get after the
radio is on for a half hour?

Its next to a big power (5Y3) tube, that's rather hot. I can place my
hand over the transformer, it feels very warm. Its not making any
burning smells,hisses etc. Will it burst into flames?

Whats the method for testing a transformer with Volt-Ohm meter? This
is first radio I've worked on that has a transformer, the others were
AA5 hot chassis sets.

Thanks! Barry F.

Bill Baka

unread,
Nov 23, 2009, 10:54:08 PM11/23/09
to

Moderately warm, about 130 to 140 degrees F.
That's a ballpark figure, since some designers pushed things a bit too
hard. It still happens today, cutting corners.

Good luck.
Bill Baka

Richard Knoppow

unread,
Nov 23, 2009, 10:58:56 PM11/23/09
to

<fleis...@comcast.net> wrote in message
news:87a91036-5b54-4c3e...@p28g2000vbi.googlegroups.com...


I've found that a lot of transformers run hot enough to
be uncomfortable to touch.


--

--
Richard Knoppow
Los Angeles
WB6KBL
dick...@ix.netcom.com


Bill Noble

unread,
Nov 24, 2009, 1:05:39 AM11/24/09
to
if you are worried, you can reduce the load on the Xformer by using a pair
of 1n4008 diodes in lieu of the 5y3, and then disconnect the filament to the
5y3 - add a series resistor to account for the lower Vfd of the diodes vice
the tube

<fleis...@comcast.net> wrote in message
news:87a91036-5b54-4c3e...@p28g2000vbi.googlegroups.com...

Nordic Breeds

unread,
Nov 24, 2009, 1:12:24 AM11/24/09
to
<fleis...@comcast.net> wrote in message
news:87a91036-5b54-4c3e...@p28g2000vbi.googlegroups.com...
> Hi all,
>
> How warm should a transformer in a 40's console radio get after the
> radio is on for a half hour?
>
> Its next to a big power (5Y3) tube, that's rather hot. I can place my
> hand over the transformer, it feels very warm. Its not making any
> burning smells,hisses etc. Will it burst into flames?


60 degrees Celsius (140 F) is the maximum surface temperature for good
design. This is the highest temperature an average person can touch and
hold his finger there, yet it is several degrees below which a physical
burn occurs. Transformers typically lose half their life for every 8
degrees C rise in temperature. Now being one of the most expensive
components to replace in old radios, it is best to keep the transformer
as cool as possible. Very few old radios had adequate ventilation. This
is why I always replace the vacuum rectifiers with silicon diodes.
Typically 10 to 15 watts can be saved by this technique depending on the
rectifier tube used. You will usually have to add a power resistor in
series with the silicon diodes to drop the voltage. The power dissipated
in this resistor is identical to the power dissipated by the tube's plate
resistance. Also, this resistor can often be bolted to the metal chassis
or otherwise positioned for good ventilation.

73, Dr. Barry L. Ornitz WA4VZQ


Alex

unread,
Nov 24, 2009, 10:15:04 PM11/24/09
to

<fleis...@comcast.net> wrote in message
news:87a91036-5b54-4c3e...@p28g2000vbi.googlegroups.com...

I would suggest to test the transformer at no load. Pull all the tubes out
and run the transformer for hours. It should run almost cool -- perhaps only
up to 5C dgrees above ambient. Touch it every now and then. You might notice
that iron is getting warmer before the copper windings pick up. Such
behaviour is normal and indicates no short turns. If the transformer is
getting noticeably warm on idle -- it is a bad sign. It is either short
turns or mains voltage is larger than it is set to.

If on idle the ranny is OK but under load gets hot, then try reduce the
load, for example, using Si diodes (with series resistor) instead of
5Y3.Check also for leaking electrolytics. These can become warm themselves
plus be adding to the total overloading.

Regards,
Alex


Pete Bertini

unread,
Nov 24, 2009, 6:19:56 AM11/24/09
to
Has the radio been recapped? Excessive current draw
can account for part of the problem if the set hasn't
been recapped.

Also, what is your mains voltage? Many early transformers
will core saturate on today's higher mains voltages.

I'd suggest using a filament transformer to drop the
mains voltage by 6 or 12 volts. It is wired autotransformer
fashion on the mains side to do perform that task.

Pete


John Goller, k9uwa

unread,
Nov 24, 2009, 7:34:31 AM11/24/09
to
In article <heftga$j5d$1...@news.eternal-september.org>, wa4...@live.com
says...

>
>
>60 degrees Celsius (140 F) is the maximum surface temperature for good
>design. This is the highest temperature an average person can touch and
>hold his finger there, yet it is several degrees below which a physical
>burn occurs. Transformers typically lose half their life for every 8
>degrees C rise in temperature.
>
> 73, Dr. Barry L. Ornitz WA4VZQ

The above is correct statement. However we are now dealing with an aged
transformer. 70+ year old insulation inside that transformer that has most
likely already been overheated a few times.

Today most of our homes have 122 to 125 volts to your wall plugs. That old
radio was designed for 115 volts. Maybe if its earlier perhaps 110.
Remember back in the old days. Electricians always called that stuff coming
out of your wall "One Ten"

As Pete mentioned at todays wall outlet levels many of these transformers
are into saturation point in the transformer design at 125 volts.

Although 140 degrees F is what that transformer was designed for today I
prefer to set all radios that leave my bench at 130 F and thats after
operating several hours in a room with ambient temperature of 72 degrees.

Notice in the above I said I SET THE TEMPERATURE of the transformer. This
is done with one of two methods. On lower tube count radios... usually up
to nine tube sets I do this with a large power resistor in the switch side
of the AC Line. Simple enough to install under the chassis away from parts
that might be heat damaged. On higher tube count sets or ones with motors
and other varying load items in them we install bucking transformers to
reduce the incoming line votlage to the set.

Its a simple matter to reduce the voltage to any old radio. The idea here
is that we are trying to restore this radio so that 70 more years from now
it will still be around for someone else to ejnjoy it.

Removing the 5Y3 and installing diodes and resistors only drops a little of
the load. Does nothing for the increased 6.3 filament voltage. And does
nothing for any transformer saturation taking place either.

On the Bench the radio is run from a metered variac and temperatures are
observed with a little 20 buck infrared temp gadget from Harbor Frieght.

Whatever that incoming voltage is when I am happy with the operting
temperture is what determines the reduction that set gets.

We do about 50 radios per year as a retirement business. I don't want the
sets coming back either under warranty or even long after the warranty is
up with fried transformers.

John k9uwa

Steve

unread,
Nov 24, 2009, 10:07:48 AM11/24/09
to
I've run into this myself a couple of times. In both
cases the transformer was drawing excess current with
no load. As someone else suggested, remove the load
(pull all tubes) and see if it still gets warm.

I'd also suggest measuring the line current with and without
a load. For most sets (9 tube or less) normal line current
will be less than 1/2 amp. With no load it should be a lot
less (0.1 amp or less).

Steve


jughead

unread,
Nov 24, 2009, 12:24:07 PM11/24/09
to
>
> Removing the 5Y3 and installing diodes and resistors only drops a little of
> the load.
>
10 watts for the filament alone is not insignifigant.
oc

Bill M

unread,
Nov 24, 2009, 2:51:43 PM11/24/09
to

And if that reduces the IR drop on the primary then you can expect
secondary voltages higher still.
Its all manageable but simply subbing the 5Y3 to alleviate its filament
current draw isn't a one stop fix.

Kenneth Scharf

unread,
Nov 24, 2009, 3:44:42 PM11/24/09
to
Is the transformer rusted? I've seen many radios where the power
transformer (as well as filter choke and audio output transformers) were
very rusty due to the radio being stored in a damp environment
(un-heated garages, etc). I've often wondered about the magnetic
properties of a rusted transformer core, and how that might effect the
operation of the transformer (probably NOT in a good way).

Pete Bertini

unread,
Nov 24, 2009, 3:58:26 PM11/24/09
to

"jughead" <oldco...@gmail.com> wrote in message
news:943ee752-89c2-4137...@k13g2000prh.googlegroups.com...

But, that isn't ten watts of dissipation in the transformer
that is eliminated. Plus, just replacing the diodes without
compensating for the increased peak surge currents
might mean no improvement is being made.
Don't shoot the messenger, but Tom W8JI gives a
good discourse here:

http://www.w8ji.com/power_transformer_stress.htm

Reducing the primary voltage is a pretty much sure
fire cure. I'd go as far as suggesting to use the old 5-volt
rectifier winding in series aid with the primary, to
further reduce the heating.

Pete k1zjh


Alan Douglas

unread,
Nov 24, 2009, 5:23:37 PM11/24/09
to
>Electricians always called that stuff coming out of your wall "One Ten"

A figure of speech. It was never an official standard.

By 1926, 12% of lamps sold were for 110V, 47% were for 115V, 35% were
for 120V, and 4% were for 125V (from the 1927 AIEE Transactions,
pp.161-215). In the 1930s those percentages would have been even
higher for 120V and 125V.

Alan

Kenneth Scharf

unread,
Nov 24, 2009, 5:44:45 PM11/24/09
to
Thomas Edison started out delivering 110 volts with his DC power plants.
In the first use of electric lighting in NY the generators were
regulated to deliver 110 volts. But this was at the generator. By the
time the current was delivered to the customer there was probably around
100 volts left after the voltage drop in the mains. Worst case, the
customer might only see 90 volts, if he lived next door to the power
plant he got the full 110 volts. Edison must have designed his lamps to
run at 100 volts nominal, with acceptable life at 110 volts.

When Westinghouse started taking over the lighting industry with his AC
system the problem of voltage drop in the mains was a thing of the past
since the final step down transformer was located less than a 1000 feet
or so from the load. Then 110 volts at the load was guaranteed.

Over the years the voltage has crept up as a way of squeezing more
capacity over the same wires.

Bill M

unread,
Nov 24, 2009, 6:22:50 PM11/24/09
to
John Goller, k9uwa wrote:

>
> On the Bench the radio is ... observed with a little 20 buck infrared temp gadget from Harbor Frieght.
>
> John k9uwa
>

Hiya John,

That sounds like a device that I need. Item Details or part number por
favor?

Thanks,
Bill

Bill Baka

unread,
Nov 24, 2009, 7:09:16 PM11/24/09
to
Kenneth Scharf wrote:
> Thomas Edison started out delivering 110 volts with his DC power plants.
> In the first use of electric lighting in NY the generators were
> regulated to deliver 110 volts. But this was at the generator. By the
> time the current was delivered to the customer there was probably around
> 100 volts left after the voltage drop in the mains. Worst case, the
> customer might only see 90 volts, if he lived next door to the power
> plant he got the full 110 volts. Edison must have designed his lamps to
> run at 100 volts nominal, with acceptable life at 110 volts.
>
> When Westinghouse
*WTF?*
started taking over the lighting industry with his(?) AC

> system the problem of voltage drop in the mains was a thing of the past
> since the final step down transformer was located less than a 1000 feet
> or so from the load. Then 110 volts at the load was guaranteed.
>
> Over the years the voltage has crept up as a way of squeezing more
> capacity over the same wires.

Your history is a bit off the mark. It was Tesla who pushed AC to
Westinghouse since Edison refused to listen to Tesla's warnings about
distributing DC. Edison was an arrogant asshole in the day.
Many of his */inventions/* were just improvements on other people's ideas.
Bill Baka

Bill Noble

unread,
Nov 24, 2009, 9:10:05 PM11/24/09
to
try this
http://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=93984

"Bill M" <radio...@geeeemail.com> wrote in message
news:hehpp...@news2.newsguy.com...

Bill M

unread,
Nov 24, 2009, 9:42:03 PM11/24/09
to
Bill Noble wrote:
> try this
> http://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=93984

That fits the bill. Rather amazing technology for 20 bucks.

I gotta add one to my trinket collection. See if it works on my fish
tanks and Thanksgiving Turkey as well.

-Bill

John Goller, k9uwa

unread,
Nov 24, 2009, 10:21:37 PM11/24/09
to
In article <hei5h...@news3.newsguy.com>, radio...@geeeemail.com
says...

Yup thats the same gadget... Brenda here I think was the one that pointed
them out a year or two ago. Same price as now. Xmas deals maybe.
John k9uwa

Alan Douglas

unread,
Nov 24, 2009, 10:39:05 PM11/24/09
to
>Thomas Edison started out delivering 110 volts with his DC power plants.
>In the first use of electric lighting in NY the generators were
>regulated to deliver 110 volts. But this was at the generator. By the
>time the current was delivered to the customer there was probably around
>100 volts left after the voltage drop in the mains. Worst case, the
>customer might only see 90 volts, if he lived next door to the power
>plant he got the full 110 volts. Edison must have designed his lamps to
>run at 100 volts nominal, with acceptable life at 110 volts.

No argument here. I'll quote part of my 10/26 post on voltage
standards again (numbers also from the AIEE paper):

"Edison's original idea was voltage multiples of 100, measured at the
load, but generators were rated to allow 15% drop in transmission,
and later 10%. In 1911 the voltages began to be measured at the
sending end, accounting for the shift to multiples of 11, and in 1914
motor voltages were standardized at 110, 220, 440, 550 and 2200V, with
an allowable variation of +/-10%. Appliances were not yet
standardized, but were often rated for a range of voltages."

Alan

Michael A. Terrell

unread,
Nov 24, 2009, 11:17:20 PM11/24/09
to


<http://www.harborfreight.com/cpi/ctaf/Displayitem.taf?itemnumber=93983>
has been on sale several times for $7.95, instead of the full price of
$19.99.


--
The movie 'Deliverance' isn't a documentary!

Michael A. Terrell

unread,
Nov 24, 2009, 11:19:00 PM11/24/09
to

I pointed them out along with a list of other useful items like the
resistor kits, and caught hell for it. Brenda wanted someone to ship one
to her. I don't know if she ever got one or not.

Nordic Breeds

unread,
Nov 25, 2009, 1:20:31 AM11/25/09
to
"Bill M" <radio...@geeeemail.com> wrote in message
news:hei5h...@news3.newsguy.com...


As a person who used to design instrumentation, and as someone who was
usually called in when an exotic temperature measurement was needed, all
I can say is view the readings of this unit with a grain of salt.
Infrared temperature measurements lose accuracy the lower the temperature
is below 200 C. Accurate room temperature measurements and those just
above room temperature with an infrared thermometer is extremely
difficult. First, you need to know the emissivity of the surface you
wish to measure. [The Harbor Freight device has no compensation or
adjustment for emissivity.] Then you need a detector that is sensitive
enough at the far infrared. Measurements at two wavelengths are often
performed so the thermometer can ratio the results to minimize (but not
eliminate) the effects of differing emissivity's of different surfaces.

If you want reasonable accuracy, stick with thermocouples or thermistors.
For higher accuracy use platinum RTD's which measure the resistance of a
thin platinum wire whose resistance varies with temperature. With some
care, the resistance of the transformer wire itself can be used. Tables
showing the resistivity of copper versus temperature are readily
available. If you need to measure temperatures in an electric field,
like a high voltage transformer or in a microwave oven of dielectric
heater, there are special fiber optic probes coated with a phosphor
(chosen for a particular temperature range). The phosphor is hit with a
pulse of ultraviolet light and the fluorescence decay rate is used to
measure temperature. These can be extremely fast response sensors, and
their accuracy is good.

In the manufacture of PET soft drink bottles, the plastic preform is
heated in an infrared oven which uses quartz-clad heaters. It is
possible to measure the preform temperature by using optical filters that
pass the particular emission wavelengths of the PET plastic. The
measurement is made against the exceptionally bright background of the
quartz heaters by also using filters that block the transmission
wavelengths of quarts.

Nordic Breeds

unread,
Nov 25, 2009, 1:32:06 AM11/25/09
to
"Kenneth Scharf" <wa2...@bellsouth.net> wrote in message
news:hehgjn$ghe$1...@news.eternal-september.org...

> Is the transformer rusted? I've seen many radios where the power
> transformer (as well as filter choke and audio output transformers)
> were very rusty due to the radio being stored in a damp environment
> (un-heated garages, etc). I've often wondered about the magnetic
> properties of a rusted transformer core, and how that might effect the
> operation of the transformer (probably NOT in a good way).


Excessive rust may mean the insulation between the transformer windings
has gotten wet and is damaged. Surface rust, however, does little
damage. Unless a significant amount of the iron has been rusted away,
the magnetic properties will be virtually unchanged.

Some very cheap transformer manufacturers even let the laminations rust a
little before assembling them as the transformer core. The thin layer of
oxidation acted to insulate one lamination from another. More reputable
manufacturers used shellac or varnish coatings on the laminations.

Pete Bertini

unread,
Nov 25, 2009, 8:57:49 AM11/25/09
to

"John Goller, k9uwa" <k9uwa...@THISarrl.netSTUFF> wrote in message
news:5p1Pm.144947$5n1.166@attbi_s21...

How well does it work on lighter colored objects? I have an infrared Fluke
(wasn't cheap) that only works on darker colored objects. Trying to
read water temp in a copper pipe is like herding cats--unless the
section of copper is painted flat black. They seem pretty close on
power transformers, but I'm now using a remote thermistor with a converter
box on a digital VOM.

Pete


Kenneth Scharf

unread,
Nov 25, 2009, 9:06:22 AM11/25/09
to
No I am quite correct. Tesla invented the AC system and sold his
patents to George Westinghouse who actually built power plants based on
them. Tesla NEVER sold ANYTHING directly to the public.

John Goller, k9uwa

unread,
Nov 25, 2009, 9:15:43 AM11/25/09
to
In article <heiibg$o37$1...@news.eternal-september.org>, wa4...@live.com
says...

>
>
>As a person who used to design instrumentation, and as someone who was
>usually called in when an exotic temperature measurement was needed, all
>I can say is view the readings of this unit with a grain of salt.
>Infrared temperature measurements lose accuracy the lower the temperature
>is below 200 C. Accurate room temperature measurements and those just
>above room temperature with an infrared thermometer is extremely
>difficult.
>
> 73, Dr. Barry L. Ornitz WA4VZQ
>
>

I just aimed my little Elcheapo Harbor Freight temp sensor at about a dozen
different items laying about on my bench. Everything from the bench to
glass to various tools on the bench. Temperature variation in the readings
was from 73.1 degrees to 73.9 degrees F. Wall Thermostat says 74 degrees.

So yes if your trying to get very accurate this gadget probably isn't what
you want... but for our uses looking at transformer temperatures and
tubes... and other items it more than close enough.

John k9uwa

jughead

unread,
Nov 25, 2009, 12:38:55 PM11/25/09
to
On Nov 24, 12:58 pm, "Pete Bertini" <radioconnect...@cox.net> wrote:

>
> > 10 watts for the filament alone is not insignifigant.
> >               oc
>
> But, that isn't ten watts of dissipation in the transformer
> that is eliminated. Plus, just replacing the diodes without
> compensating for the increased peak surge currents
> might mean no improvement is being made.
> Don't shoot the messenger, but Tom W8JI gives a
> good discourse here:
>
> http://www.w8ji.com/power_transformer_stress.htm
>
> Reducing the primary voltage is a pretty much sure
> fire cure. I'd go as far as suggesting to use the old 5-volt
> rectifier winding in series aid with the primary, to
> further reduce the heating.
>

Thanx for postin' that tutorial, Pete. So many of us have taken the
"10 watts saved" axiom as gospel forever without stopping to think it
through. Your suggestion of using the 5V winding for bucking is cool.
oc(Bill)

jughead

unread,
Nov 25, 2009, 1:23:24 PM11/25/09
to
On Nov 25, 9:38 am, jughead <oldcoot7...@gmail.com> wrote:

> Your suggestion of using the 5V winding for bucking is cool.
>

Er, I guess it wouldn't eggzackly be "bucking" but adding more turns
to the primary, right? (:

Pete Bertini

unread,
Nov 25, 2009, 2:52:16 PM11/25/09
to

"jughead" <oldco...@gmail.com> wrote in message
news:0e6bf884-49b9-4b0c...@u18g2000pro.googlegroups.com...

Yea, I was called on that once :). Adding the winding so it is in phase
with the primary. Increasing the # of turns on the primary winding,
thus decreasing turns ratios between primary and secondaries.

Since the 5 volt winding is designed to handle the DC voltage,
it should have ample insulation for use on the AC line side as well.
Only nitpick I can think of is doing so on those fancy transformers
with built in Faraday shields.

Pete


nesesu

unread,
Nov 25, 2009, 3:02:37 PM11/25/09
to
On Nov 25, 5:57 am, "Pete Bertini" <radioconnect...@cox.net> wrote:
> "John Goller, k9uwa" <k9uwaREM...@THISarrl.netSTUFF> wrote in messagenews:5p1Pm.144947$5n1.166@attbi_s21...
>
>
>
>
>
> > In article <hei5h30...@news3.newsguy.com>, radioex...@geeeemail.com

> > says...
>
> >>Bill Noble wrote:
> >>> try this
> >>>http://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=93984
>
> >>That fits the bill.  Rather amazing technology for 20 bucks.
>
> >>I gotta add one to my trinket collection.  See if it works on my fish
> >>tanks and Thanksgiving Turkey as well.
>
> >>-Bill
>
> > Yup thats the same gadget... Brenda here I think was the one that pointed
> > them out a year or two ago. Same price as now. Xmas deals maybe.
> > John k9uwa
>
> How well does it work on lighter colored objects?  I have an infrared Fluke
> (wasn't cheap) that only works on darker colored objects. Trying to
> read water temp in a copper pipe is like herding cats--unless the
> section of copper is painted flat black. They seem pretty close on
> power transformers, but I'm now using a remote thermistor with a converter
> box on a digital VOM.
>
> Pete- Hide quoted text -
>
> - Show quoted text -

The Fluke IR thermometer probe used to come with a sheet of black
stick on dots that had a very good emissivity. If the item is of poor
emissivity you just stuck one on and then the reading was fairly
accurate. One can also use an HP [Fluke?] thermal probe [contact type]
to measure. We used to use the contact probe to verify the radings on
the IR probe before using the IR to measure the unit while operating
[usually at high voltage and powers where using the contact probe was
unsafe]. We found that black plastic electrical tape with a dull
outside surface [not shiny] worked just as well. We just cut 3/4"
squares and stuck them onto everything we wanted to measure. That also
gives easily repeatable targets.

Neil S.

Pete Bertini

unread,
Nov 25, 2009, 3:07:04 PM11/25/09
to

"nesesu" <neil_su...@telus.net> wrote in message news:2804caf0-3f24-

The Fluke IR thermometer probe used to come with a sheet of black
stick on dots that had a very good emissivity. If the item is of poor
emissivity you just stuck one on and then the reading was fairly
accurate. One can also use an HP [Fluke?] thermal probe [contact type]
to measure. We used to use the contact probe to verify the radings on
the IR probe before using the IR to measure the unit while operating
[usually at high voltage and powers where using the contact probe was
unsafe]. We found that black plastic electrical tape with a dull
outside surface [not shiny] worked just as well. We just cut 3/4"
squares and stuck them onto everything we wanted to measure. That also
gives easily repeatable targets.

Neil S.

Black targets would be neat! My unit came from Graybar
Electric, and was intended for electricians to aid searching
for overheating breakers. It looked like a neat toy, so I
bought it on a lark. No targets with it, unfortunately.

I also use mine to check supply and return temps on my boiler.
I'll have to do a search for a source for those targets.

Pete


Kenneth Scharf

unread,
Nov 25, 2009, 3:43:50 PM11/25/09
to
Well you can look at it either way. You ARE changing the turns ratio of
the entire transformer by putting the 5v winding in series with the
primary and thereby increasing the number of turns on the primary.
You are also sorta kinda creating an auto transformer to lower the
effective voltage seen across the actual primary. 6 of 1 or a half
dozen of the other. The math works out the same either way. Note that
the 5 volt winding must be able to handle the required primary current,
I doubt that the radio requires more than 220 watts of power (110 volts
X 2 amps), so you should be OK there.

Michael A. Terrell

unread,
Nov 25, 2009, 4:15:47 PM11/25/09
to


It's simple to use a lookup table to correct the reading.
Microprocessors are well under a buck these days, even in single
quantities. I also have a DVM meter with a thermocouple probe and every
time I compared readings there was less than one degree difference. No
one claimed that either are traceable to NIST.

Heywood Jablowme

unread,
Nov 25, 2009, 5:15:08 PM11/25/09
to
I could be way off my game here but


has anyone considered this just might be a 25 cycle transformer ?? at a
higher freq I would think it would runner hotter.

<fleis...@comcast.net> wrote in message
news:87a91036-5b54-4c3e...@p28g2000vbi.googlegroups.com...
> Hi all,
>
> How warm should a transformer in a 40's console radio get after the
> radio is on for a half hour?
>
> Its next to a big power (5Y3) tube, that's rather hot. I can place my
> hand over the transformer, it feels very warm. Its not making any
> burning smells,hisses etc. Will it burst into flames?
>
> Whats the method for testing a transformer with Volt-Ohm meter? This
> is first radio I've worked on that has a transformer, the others were
> AA5 hot chassis sets.
>
> Thanks! Barry F.


Pete Bertini

unread,
Nov 25, 2009, 5:46:03 PM11/25/09
to

"Heywood Jablowme" <SOM...@SOMEWHERE.COM> wrote in message
news:hekat...@news6.newsguy.com...

>I could be way off my game here but
>
>
> has anyone considered this just might be a 25 cycle transformer ?? at a
> higher freq I would think it would runner hotter.
>

Would be the opposite. The lower frequency transformer has more iron
and reactance. It should run fine at the higher frequency. Trying to run
a transformer at a lower frequency, like an aircraft 400 Hz transformer,
at 60 Hz would be disasterous.

Pete


Bill Baka

unread,
Nov 25, 2009, 8:27:11 PM11/25/09
to
Kenneth Scharf wrote:
<snipped myself>

> No I am quite correct.
As to the marketing part of things, yes, but it was Tesla who worked
with them to get AC going.

Tesla invented the AC system and sold his
> patents to George Westinghouse who actually built power plants based on
> them. Tesla NEVER sold ANYTHING directly to the public.

I wasn't thinking 'sold' but rather who the driving force was. If Edison
had his way we would be way behind on development. It is very well known
that Tesla went to Westinghouse after giving up on Edison and his "I'm
right, everybody else is wrong." attitude.

Bill Baka

Nordic Breeds

unread,
Nov 25, 2009, 11:44:28 PM11/25/09
to
"John Goller, k9uwa" <k9uwa...@THISarrl.netSTUFF> wrote in message
news:j_aPm.149248$la3.23943@attbi_s22...

Looking at black transformers yes, but looking at a glass tube trying to
get the seal temperature, forget it. Depending on the actual wavelength
used, you may see the plate, or the filament poking out from the cathode,
or something else entirely. The old Tempilaq crayons (available from
Omega) are still used. There are now stick-on metal alloys that melt at
known temperatures. They go from dull to shiny when the alloy melts.

The Harbor Freight device assumes an emissivity of 95%, and there is no
adjustment for this. Eimac discussed the use of Tempilaq crayons in "The
Care and Feeding of Power Grid Tubes".

For those not familiar with Tempilaq crayons, they are colored waxes that
are rubbed on a surface to be measured. At room temperature, they leave
a dull streak. When the melting temperature of the crayon is reached,
the streak turns clear or shiny.

Nordic Breeds

unread,
Nov 26, 2009, 12:04:08 AM11/26/09
to
"Michael A. Terrell" <mike.t...@earthlink.net> wrote in message
news:uvmdnU0r0PgQA5DW...@earthlink.com...

>
> It's simple to use a lookup table to correct the reading.
> Microprocessors are well under a buck these days, even in single
> quantities. I also have a DVM meter with a thermocouple probe and every
> time I compared readings there was less than one degree difference. No
> one claimed that either are traceable to NIST.
>
> --
> The movie 'Deliverance' isn't a documentary!


A lookup table is good only if you know the actual emissivity of a
surface. The
following is a quote from OptoTherm's website:

" Emissivity in the Infrared
Emissivity Table
Emissivity is a measure of a material�s radiating efficiency. An
emissivity of 1.00
implies that the material is 100% efficient at radiating energy. An
emissivity of
0.20 implies that the material radiates only 20% of that which it is
capable of
radiating.

Tables of emissivity values are only approximated values for real
materials. A
range of emissivity values is usually given for many materials whose
emissivity can
be affected by surface roughness or finish. Additionally, thin sheets of
material
such as plastics may be semi-transparent in the infrared and therefore
have
reduced emissivity.

To optimize the surface temperature measurement of a material:

Avoid reflections by shielding the material from surrounding high
temperature
objects.
For semi-transparent materials such as plastic film, assure that the
background
is uniform and lower in temperature than the material.
Conduct the measurement perpendicular to the material�s surface
whenever
the emissivity is less than approximately 0.90. In all cases, do
not exceed
angles greater than 30 degrees from perpendicular"

Polished metals have quite low emissivities. OptoTherm publishes a table
of
emissivity ranges for many ,aterials. .

http://www.optotherm.com/emiss-table.htm

Kenneth Scharf

unread,
Nov 26, 2009, 10:59:56 AM11/26/09
to
Well Tesla was quite the showman. He often demonstrated that he could
allow 1,000,000 volts of HF AC to pass though his body without any harm.
This trick was possible because the current was very low and the
frequency was high enough not to be felt, but low enough not to burn.
(probably somewhere around 10-100 KHZ which was HF back then!).

Kenneth Scharf

unread,
Nov 26, 2009, 11:01:32 AM11/26/09
to
except that there ARE some transformers made that could handle such a
wide frequency range (50-400hz). No doubt a VERY special core design.

Pete Bertini

unread,
Nov 26, 2009, 11:18:35 AM11/26/09
to

"Kenneth Scharf" <wa2...@bellsouth.net> wrote in message
news:hem8om$bte$3...@news.eternal-september.org...

In general, since the model number is unknown...

I should note, if you do use a radio with a 25 Hz transformer on 60 Hz, you
might need to change the filter capacitor values.

The parts may have been optimized for 50 Hz filtering, not 120 Hz (assuming
full
wave rectification.) The value of the input filter is especially critical.
The original value, if too great, will unduly stress both the rectifier and
transformer.
If there is an input filter capacitor, 10 mFd is a good starting point for
a 60 Hz
fullwave transformer.

Pete


Michael A. Terrell

unread,
Nov 26, 2009, 1:42:37 PM11/26/09
to

Nordic Breeds wrote:
>
> "Michael A. Terrell" <mike.t...@earthlink.net> wrote in message
> news:uvmdnU0r0PgQA5DW...@earthlink.com...
> >
> > It's simple to use a lookup table to correct the reading.
> > Microprocessors are well under a buck these days, even in single
> > quantities. I also have a DVM meter with a thermocouple probe and every
> > time I compared readings there was less than one degree difference. No
> > one claimed that either are traceable to NIST.
>
> A lookup table is good only if you know the actual emissivity of a
> surface. The following is a quote from OptoTherm's website:


Thanks, but I am quite familiar with emissivity. We were talking
about reading power transformer temperatures, not tubes or anything
else. If we were, I would have said the same thing. I have worked in a
metrology lab, as well.

Emissivity and contactless IR thermometers have been discussed to
death on the news:sci.electronics.design newsgroup.

> " Emissivity in the Infrared
> Emissivity Table
> Emissivity is a measure of a material�s radiating efficiency. An
> emissivity of 1.00 implies that the material is 100% efficient at
> radiating energy. An emissivity of 0.20 implies that the material
> radiates only 20% of that which it is capable of radiating.
>
> Tables of emissivity values are only approximated values for real
> materials. A range of emissivity values is usually given for many
> materials whose emissivity can be affected by surface roughness or
> finish. Additionally, thin sheets of material such as plastics
> may be semi-transparent in the infrared and therefore have
> reduced emissivity.
>
> To optimize the surface temperature measurement of a material:
>
> Avoid reflections by shielding the material from surrounding high
> temperature objects.
> For semi-transparent materials such as plastic film, assure that
> the background is uniform and lower in temperature than the material.
> Conduct the measurement perpendicular to the material�s surface
> whenever the emissivity is less than approximately 0.90. In all cases,
> do not exceed angles greater than 30 degrees from perpendicular"
>
> Polished metals have quite low emissivities. OptoTherm publishes a table
> of emissivity ranges for many ,aterials.
>

> http://www.optotherm.com/emiss-table.htm

Michael A. Terrell

unread,
Nov 26, 2009, 1:43:25 PM11/26/09
to

Heywood Jablowme wrote:
>
> I could be way off my game here but
>
> has anyone considered this just might be a 25 cycle transformer ?? at a
> higher freq I would think it would runner hotter.


No, it would run cooler.

Nordic Breeds

unread,
Nov 26, 2009, 7:47:19 PM11/26/09
to
"Kenneth Scharf" <wa2...@bellsouth.net> wrote in message
news:hem8om$bte$3...@news.eternal-september.org...


Not really/ I have several pieces of military test equipment with
transformers rated to operate from 50 Hz to 1000 Hz. My TS-413 signal
generator, for example, says:

POWER SUPPLY: This test set requires, for satisfactory operation, a
power source supplying 105 to 125 volts at 60 to 1600 cycles per
second.

The URM-25D, very popular with ham radio Boatanchor restorers requires
115 volts +/- 10% at 50 to 1000 cycles per second. The Hewlett Packard
608D RF signal generator and the HP-200 audio signal generators require
115(/230) volts +/-10% at 50 to 1000 Hz. The power transformers in these
pieces of equipment are hardly exotic.

Probably one of the best books on transformers I have seen is "Magnetic
Circuits and Transformers - A First Course for Power and Communications
Engineers". It was first published in the early 1940's by the MIT Press.
The 1977 edition is still available as a 'print-on-demand' book for $75.
Used copies are often available at slightly lower costs.

Dr. Barry L. Ornitz WA4VZQ (45 years a ham operator as of this
month)


Bill Baka

unread,
Nov 26, 2009, 9:47:11 PM11/26/09
to
A solid arc from a marine radio using tubes will wake one up. I did it
and got probably 600 to 1,000 volts of RF into my finger. I jumped back
but my finger still kind of hurt after 2 days. Someone told me that R.F.
tends to make a bee-line for any bones.
Sounds fishy but it was my finger bone that hurt, or sure seemed like it.
At any rate we should all be damned glad we had Tesla to police Edison's
screw ups.
Bill Baka
0 new messages