|[AMRadio] Responses to: [Amps] Drying out HV transformers / Chokes|
w8hrq at lemleys.org
Sun Sep 18 21:54:41 EDT 2005
Wow, Thanks to everyone who answered my posting, both to the lists, and
I received lots of interesting ideas on how to solve my problem of
drying out transformers which have been exposed to moisture. Thanks to
all of you who took the time to respond. I really appreciate the help.
Since I posted my original query to 3 different mailing lists, and I
also received direct mail responses, I thought I would summarize what I
have received so far, for the benefit of anyone else who may run into
The ideas fall into two main categories. Internal Self-Heating, using a
current flowing in one of the windings to heat the transformer from the
inside out, and, External Heating, applying heat from some external
source until the unit has dried from the outside in.
Obviously from the accounts below, both methods are known to work, so I
am trying to make the best choice for my circumstances. Someone else
may have much different priorities. So far, I am leaning toward the
Internal Self-heating route, because I think I can control the process
more closely, and still get the job done rapidly. Speed is important
for me, as I am trying to get this thing on the air for the upcoming
top-band season. I am currently thinking that the best method for me
might be to use AC from the mains thru a Variac and the secondary, with
an ammeter on the primary to measure current.
For those of you with experience with this, I have a question or two.
First, Steve Hobensack, suggested that I use DC instead of AC, as did
Bob Bruhns and Bill Hawkins, however, they also pointed out that this
would have a tendency to magnetize the core, and even suggested a couple
of fixes for that problem. However, I would rather not have to deal
with the magnetization problem at all. What advantage is there to using
DC for this purpose?
My second issue is that of sealing the newly dried transformer once I
have finished. Different postings mentioned variously HV varnish,
polyester laminating resin, polyester varnish, etc. What is the actual
terminology and can you recommend sources?
Last question, where can I get access to a "megger" (without having to
invest a whole lot of money for a one or two time use)? Or, can they be
had inexpensively somewhere?
Ok, that's it for my new questions, the following is the summary of the
received responses, along with credit to those folks who volunteered
information. Again, thanks to you all for the input, and as usual, I
learned a lot in the process.
Internal Self-heating - All of these suggestions centered on shorting
one or the other windings of the transformer, and applying power to the
other, until sufficient current was drawn to cause the transformer to
begin to heat from the inside. There were multiple variations on this
theme, including the following:
Steve Thompson [g8gsq at ic24.net]
Self-heating? An untried idea - short one of the windings, and
apply a little voltage to the other with a variac and let the
thing warm up gradually, increasing the voltage until it sits at
50C/120F for a few days. My guess would be to short the primary and
apply volts to the secondary.
Dfmich at aol.com
I have not done this but had a "brainstorm". How about shorting
the secondary of the xfmr with a 1 ohm resistor. Apply AC to the
primary with a variac and adjust the volts in to get rated
current in the secondary as indicated by the voltage drop across the 1
ohm resistor. (you'd have to pick a resistor wattage to withstand the
current/dissipation). The primary voltage will probably be fairly low;
and the voltages in general will be low so no arcing should occur.
Let the whole shebang set for say a week. hopefully it will get warm
and the heat will chase out the moisture. Your can feel the thing to
determine if any seemingly significant heat is being generated. If
not, try something else!
Dave Brown [tractorb at ihug.co.nz]
If you know the secondary current rating or can guess it
approximately then you are all set. Disconnect all connections to the
transformer- both primary and secondary. Short the secondary with an
AC ammeter that will be used to indicate the rated current flowing
in the secondary. Run the primary up on a variac - CAREFULLY! - until
you see the rated current indicated on the ammeter. It's not all
that critical -as long as the current is around the right value it
will do the business. In this configuration you are way below the
overall VA rating for the xfmr as a whole so a bit of extra
dissipation in the secondary isn't going to matter. A few percent
high or low is fine. The secondary resistance is used to dissipate
power at the rated current in the secondary and thus heat up the
transformer from the inside out. Then just leave everything like
that for a week, or however long you want to. Don't disconnect
anything without first winding the variac right down!
Steve Hobensack [stevehobensack at hotmail.com]
It may be hard to pump enough ac current through a high voltage
secondary that comes from a variac. You can use a variable source of
DC, crank it up until the 100 pound piece
of iron will take in about 50 watts. Fifty watts is quite hot.
For smaller pieces of iron, use only a few watts.
Bill Hawkins [bill at iaxs.net]
You could try self-heating by dissipating some DC watts in the
windings. The DC will magnetize the core, so run AC from a series
light bulb through the winding until the bulb dims. If the bulb
doesn't dim then reduce the magnetization by using half the DC current
in the other direction for a few seconds. Repeat swapping using half
the previous current until a 100 watt bulb is dim on AC. I've never
tried this, but that's what ought to happen.
How do you know when it's dry? Use a 500 volt megger to the case
and stop when the megohms won't reduce any more.
Bob Bruhns [bbruhns at erols.com]
Low-voltage DC on the windings would do it. It would be tricky
to figure a current that would produce enough heat without toasting
anything, though. Probably something around the maximum operating
current would do it, over time. I would put the primary and secondary
windings in series, and heat the whole core. You might try to arrange
the connections to approximately cancel the DC flux in the core. Put
the core in a blanket so the heat builds up, and monitor its
temperature. My guess is it will take days to build up the heat. You
want to heat it up so it is warm to hot to the touch, (maybe 140F /
60C), but you don't want to scorch it.
John Lyles [jtml at losalamos.com]
You could put a short on the secondary terminals, and run up the
primary voltage with a variac, to where you get the full current
flowing in the short. Voltage will be, of course, low, and this will
heat the transformer up without overloading it. Be sure to not exceed
the rated DC current, and maybe to be safe, run it only 50-75% of
that. Other than this, you could put a plywood or polyethylene sheet
tent around it, and heat it with a radiant heater of any type. All you
need is 120 deg F or so for so many days.
John Lawson [jpl15 at panix.com]
I second this, with a slight caveat - I've used this technique
over the years for transformers and chokes that have been in damp
places, out in the weather, (or in a few cases actually
submerged) - however I terminate the secondaries in some appropriate
resistance - sometimes various lamps work nicely for this -and then use
a Variac on the primary, and use plenty of patience!
If there's a lot of water, you have to take it very slow as it's
sure to arc no matter what you do - this causes insulation damage and
super-heated steam in the windings. But this is by far the best
method, since it heats 'from the inside out' - all the major utilities
use this when drying out Really Big Iron - even to desiccate the
insulating oil. But I don't think I'd short the secondaries - this
might tend to concentrate excessive currents at any 'weak spots'.
Gene Bigham [jbigham2 at kc.rr.com]
Variac in the primary, slowly bring up to full potential over
time. I did this with a huge Gates transformer that went under
water during a flood.
R. Measures [r at somis.org]
I heat large xfmrs by shorting the primary with an AC ampere
meter and connecting the secondary to a variac that is connected
to the electric mains. I increase the sec. V until I arrive at
the normal pri. current. I put a folded in half bath towel over
the unit to hold in heat. After a few hours, the xfmr should be
almost too hot to touch. Continue for a few days. At this point,
the xfmr should be dried out. At this point, polyester laminating
resin can be poured into the windings to fill the air spaces. When
the resin hardens, the
xfmr will not absorb moisture, it will be able to get rid of
internal heat better, and the insulation will be improved.- note
- xfmr potting is covered in: http://www.somis.org/D-amplifiers3.html
External Heating - These methods are based on applying heat from the
outside of the transformer or choke until it is sufficiently dried.
Several approaches: Custom oven for the purpose; use the heat from the
sun; simulate the sun (sun-lamp); and electric space heaters. Again, I
received several variations on the theme, including success stories.
Larry Will [lhwill at verizon.net]
Here is the section in my article from ER Magazine on rebuilding
an RCA BTA-1R1 on the homemade oven I used with great success.
All you need is a thermostatically controlled hotplate a
thermometer and a hood.
A check of all the iron with a 1000V 1000 megohm ohmmeter showed
excessive leakage (less than 1000 megs) in the modulation reactor, the
driver plate transformer, and the control transformer. The control
transformer makes 110VAC for relays and lamps from the 240V input.
The smaller transformers were baked in the kitchen oven at 140 degrees
F for 5 hours. For the very heavy 50 henry modulation reactor, I
fashioned a homemade "oven" using a thermostatically controlled single
burner hotplate and a hood made from an old water heater jacket
(Figure 1). This allowed me to cook the transformer out in the garage
near the rig. An oven thermometer allowed me to set the
oven temperature to 140 degrees F and after 25 hours of
"baking", the leakage was cured. All of these transformers and chokes
were then dipped in transformer varnish to re-seal the winding
from new moisture incursion.
John Coleman ARS WA5BXO [wa5bxo2005 at pctechref.com]
I have a modulation XFMR that was under muddy water for 6
It was let to stand in the sun for about 3 months under shed. I
then measured 1000 ohms from any winding to another and to ground. I
took the end caps off exposing the coil, rust and mud. Washed it out
some more and decided to put it on the shelf in a un-air conditioned
office that no one used, where the temp would reach 120-140 every day
in the summer. I was thinking I would rewind it some day. But alas, 6
months later all the readings were infinite. To be safe it was mounted
on a wooden shelf in the XMTR and has been working for the last 7-8
years. Modulator is four 813s with 2500 volts on the plates,
800 on the screens and about -100 on the grids. This modulator
will pump out some PTP voltage. I was really surprised at the come
HAROLD B MANDEL [ka1xo at juno.com]
Do you live in a sunny area? Take the xfmr and put it in over a
"parabolic" reflector made out of cardboard with tinfoil on it,
focusing the sun's energy on the iron, which is sitting on some 2x4's
so the underside is raised. A day in the sun with this will dry it
out well. Get some HV varnish and goop it up real well after the
iron is real warm. Turn it over and goop it up from the bottom up
after the top layer dries some. A few layers of HV varnish will soak
in to the innards with the core and winding nice and warm.
Joe A. Taylor [n4nas1 at hotmail.com]
I'm sure may be too simplistic to work but..... why not put the
xformer in the trunk of a car and leave it there, outside in the sun,
for a week or month. Here in KY that works for most everything.
I've dried apples this way ;-)
Electric Space Heaters
Bob Bruhns [bbruhns at erols.com]
Another thought is to get a few electric space heaters and point
them at the unit, and let them heat it up. Again, monitor its
temperature. With this kind of heating power, you really need to
watch it closely
Will Matney [craxd1 at verizon.net]
You can use an electric space heater too. One like the old
"milk-barn" types works well. Just set the transformer up with its
layers open to the heater about 6-8 inches away. You may have to turn
it once exposing the other open side. I'd advise doing this
though where nothing could be caught on fire for some odd reason as
this takes a while. The heater itself should pose no problems if it
has an active tip-over switch. Let it cook for a few days and it
should be as dry as a bone. It needs to be pretty hot to the touch
but not to where it would actually burn the fish paper. Baking
temperature in ovens is around 300-350 degrees but remember
that's for wet varnish. A through- temp of around 120-150 degrees
should do it. The main thing is to dry out the paper layer insulation
and any moisture that may be deep inside.
k7fm [k7fm at teleport.com]
Put a heater on it. That will warm up the iron, which will warm
all else. Leave the heater on for a day or so. Even a close
floodlight would help.
Richard Dillman [ddillman at igc.org] - Restoring KPH experience
Our solution was to remove them from the transmitters and bake
them under sun lamps for extended periods - a week or sometimes
several weeks. We monitored progress with a megger. Eventually
those transformers came back to life and they are in use today.
mikea [mikea at mikea.ath.cx]
I thought about sunlamps, and that seemed like a reasonable
approach. Then I thought about electric blankets. That's cheaper,
involves less energy, and so will take longer, but it's easy to do:
just wrap Mr. Xfmr up in one or two blankets set to high, and wait a
while. He'll be all dry and toasty warm in a week or three.
StephenTetorka at cs.com
How about a 500 watt light bulb in a covered box?
Sealed system with desiccant
Morris Dillingham [mdilli at trip.net]
How about placing them in a sealed container with desiccant?
The desiccant will absorb the moisture. I can't tell you where to
find the desiccant.
More information about the AMRadio mailing list
This page last updated 17 Oct 2017.