Much has been posted recently about the Heath Cantenna and the use of
mineral oil to help increase its short-term power rating. While doing a
search on a slightly similar subject, I ran across an interesting fact
about the resistors used in the Cantenna which conveniently explained a
problem I had seen with my unit.

The original non-inductive resistors used in Heath's Cantenna were made by
Carborundum Corporation's Electric Products Division (which was purchased
by Kanthal Globar Elektrowarme GmbH in 1993). Heath, as a company, was
known for reasonably good engineering, but it was never known for using
quality components. After all, the company started its electronic kit
business using surplus components. My Cantenna (HN-31) was a good example.

I purchased my Cantenna kit in the late 1960's and used transformer oil
provided by the local electric cooperative to fill the paint can. It was
much later that I learned that the oil was contaminated with
polychlorinated biphenyls. When new, the resistance was close to 50 ohms,
but over several years the resistance gradually rose to approximately 100
ohms. When checked with my VSWR meter, the dummy load showed a consistent
2:1 VSWR. I thought I had "cooked" the resistor by using high power for an
extended period, never thinking that the problem was inherent with the
resistor itself.

It turns out that Heath had supplied a type A or AS resistor rather than
the type SP which should have been used. I discovered this accidently in a
search of QST articles. In the January 1989 Technical Correspondence
column, there is a quote from the Carborundum catalog discussing the use of
a heat transfer fluid to increase the power handling capability of these
non-inductive resistors:

"Unless the resistors are protected by a coating that the fluid cannot
permeate, such as an epoxy, these fluids cause the resistance of the
Type AS
resistors to increase. With some silicones, this increase is as little
as
10%; with some mineral oils this can be as much as 100%. Generally,
the
resistance will rise as the fluid permeates the resistor body, and it
will
finally stabilize..."

The Type SP resistors are already coated at the factory, and these should
have been the ones Heath used. Note, however, that the epoxy coating
reduces the heat transfer from the silicon carbide resistor to the oil to
some extent. It also reduces the maximum voltage rating by as much as 50%.

In an earlier post, I had suggested the use of Bourns power RF flanged chip
termination resistors stocked by Digi-Key (a 250 watt unit selling for
$27.50). I have since learned that they also stock 800 watt units selling
for $78 at the time of this writing. These are good to 1 GHz. They do
contain Beryllium Oxide so be very careful when installing these. A data
sheet from Bourns may be found at
http://www.bourns.com/data/global/PDFs/CHF190104CBF.pdf. These devices
require careful attention to heat sinking, even if they are approximately
1" x 2" in size.

73, Dr. Barry L. Ornitz WA4VZQ
[transpose numbers to reply]

"Jeff Liebermann" wrote in message
...
On Fri, 23 Jan 2009 00:41:46 -0500, "Barry L. Ornitz"

wrote:
"Unless the resistors are protected by a coating that the fluid cannot
permeate, such as an epoxy, these fluids cause the resistance of the Type
AS
resistors to increase. With some silicones, this increase is as little
as
10%; with some mineral oils this can be as much as 100%. Generally, the
resistance will rise as the fluid permeates the resistor body, and it
will
finally stabilize..."

For a Carborundum resistor, if the mineral oil causes the resistance
to increase, I would guess(tm) that it would corrode or otherwise
remove material from the resistive element. That might [make] it a one
way
trip, where subsequent solvent cleaning, heating, and drying would
probably not return the resistor to its original value. Doubling the
value suggests that a considerable amount of material was corroded
away.

If the resistor material were [like] carbon composition [resistors], the
mineral
oil would act as an insulator between carbon grains, reducing the number
of points of contact, and also increasing the resistance. However,
this should be recoverable by cleaning and baking.

Are either of these mechanisms probable?

From what I know about the original Acheson Process and the later Lely and
Modified Lely Processes for making silicon carbide, and about subsequent
sintering processes, I suspect the true mechanism is somewhere in between
the two extremes that Jeff suggests, but the second mechanism is more
likely. Silicon carbide and graphite are quite inert to organic solvents.
It would even take something like hydrofluoric acid to attack the material
at such low temperatures. Knowing exactly how the material is sintered
would help with the explanation. Three main methods of sintering silicon
carbide are used. One uses glass frit or metal as a bonding agent. The
second uses graphite and silicon metal which is reaction bonded to the
silicon carbide grains. Finally boron carbide is used as a sintering aid
for very high temperature applications. Since sintering only bonds a
composite material in discrete points (as opposed to fully melting the
material), an insulating liquid could diffuse between grains and increase
the bulk resistance.

Have you tried to "repair" the load resistor?

No. Once I discovered that the transformer oil I had been given was
contaminated with PCB's, I disposed of the Cantenna. The PCB containing
oil was burned in an EPA-rated incinerator, and the Carborundum resistor
was sent to a hazardous waste landfill. I had the original HN-31, so I
rinsed the metal parts with solvent that was burned with the oil. I never
tried washing the resistor, and I was afraid to vaporize the trapped
mineral oil because of the PCB contamination.

By this time, I had purchased a high power Bird termination load at a
hamfest for $5. It had originally been used as a termination for a TACAN
system and was marked accordingly with a metal plate stating that it was
rated for a kilowatt in the frequency range of 900 to 1250 MHz. The seller
had dropped the price in increments all day and yet hams were not
interested thinking it would only work in that frequency range. I realized
that this was a TEM and not a filled waveguide load and I got an
exceptional deal. The seller had not bothered to read the resistance with
an ohmmeter! Not wanting to pay Bird's high prices for a Type-N adapter, I
made one myself.

73, Dr. Barry L. Ornitz WA4VZQ
{transpose the digits to reply}

NoSPAM wrote:
Much has been posted recently about the Heath Cantenna and the use of
mineral oil to help increase its short-term power rating. While doing a
search on a slightly similar subject, I ran across an interesting fact
about the resistors used in the Cantenna which conveniently explained a
problem I had seen with my unit.

The original non-inductive resistors used in Heath's Cantenna were made by
Carborundum Corporation's Electric Products Division (which was purchased
by Kanthal Globar Elektrowarme GmbH in 1993). Heath, as a company, was
known for reasonably good engineering, but it was never known for using
quality components. After all, the company started its electronic kit
business using surplus components. My Cantenna (HN-31) was a good
example.

I purchased my Cantenna kit in the late 1960's and used transformer oil
provided by the local electric cooperative to fill the paint can. It was
much later that I learned that the oil was contaminated with
polychlorinated biphenyls. When new, the resistance was close to 50 ohms,
but over several years the resistance gradually rose to approximately 100
ohms. When checked with my VSWR meter, the dummy load showed a consistent
2:1 VSWR. I thought I had "cooked" the resistor by using high power for
an extended period, never thinking that the problem was inherent with the
resistor itself.

It turns out that Heath had supplied a type A or AS resistor rather than
the type SP which should have been used. I discovered this accidently in
a search of QST articles. In the January 1989 Technical Correspondence
column, there is a quote from the Carborundum catalog discussing the use
of a heat transfer fluid to increase the power handling capability of
these non-inductive resistors:

"Unless the resistors are protected by a coating that the fluid cannot
permeate, such as an epoxy, these fluids cause the resistance of the
Type AS
resistors to increase. With some silicones, this increase is as little
as
10%; with some mineral oils this can be as much as 100%. Generally,
the
resistance will rise as the fluid permeates the resistor body, and it
will
finally stabilize..."

The Type SP resistors are already coated at the factory, and these should
have been the ones Heath used. Note, however, that the epoxy coating
reduces the heat transfer from the silicon carbide resistor to the oil to
some extent. It also reduces the maximum voltage rating by as much as
50%.

In an earlier post, I had suggested the use of Bourns power RF flanged
chip termination resistors stocked by Digi-Key (a 250 watt unit selling
for $27.50). I have since learned that they also stock 800 watt units
selling for $78 at the time of this writing. These are good to 1 GHz.
They do contain Beryllium Oxide so be very careful when installing these.
A data sheet from Bourns may be found at
http://www.bourns.com/data/global/PDFs/CHF190104CBF.pdf. These devices
require careful attention to heat sinking, even if they are approximately
1" x 2" in size.

73, Dr. Barry L. Ornitz WA4VZQ
[transpose numbers to reply]

I found my HN31 had the same symptom though, I don't know what kind of
coolant was used (the unit came to me used, in the 70s, with no info). I've
since refurbished it w/ a replacement resistor (Type SP) and non-PCB xfmr
oil (courtesy of my local power company). That was about a year ago.

Recently, I came upon a few dozen RF Power Labs (now Anaren) RFP400-50R
(400W, 50-ohm) flange-mount resistors, some good-sized heatsinks, and 120mm
x120mm x 100cfm tubeaxial fans (scrapped parts from my employer). I
arranged four resistors in series-parallel on two heatsinks, with the
heatsink mounting surfaces face-to-face, and the fins vertical. A fan blows
air upward thru the heatsink fins. I have an LM35DT centigrade temperature
sensor affixed to one of the heatsinks. Feeding 1100W at 40.68MHz into the
load produced a stabilized temperature of about 55°C. The resistors are
rated to 100°C at 100% of rated power. Me thinks they'll be fine for
Amateur use.

73,
Bryan WA7PRC

I purchased a Cantenna from Heathkit in about 1961 or 62. I remember
finding a gallon of mineral oil in a drug store at a low price. (It must
have been a really low price to impress me at the time.) It has been in the
Cantenna ever since, and still works well. I do not use it much, but I
sometimes run 100w or about 700w (depending on whether the AL-80B is being
used) without problems. SWR indicated in the 1.1 or 1.2 range. It was a
great investment.

Bill
W2WO

On Jan 24, 1:36*pm, "Bill Ogden" wrote:
I purchased a Cantenna from Heathkit in about 1961 or 62. *I remember
finding a gallon of mineral oil in a drug store at a low price. *(It must
have been a really low price to impress me at the time.) *It has been in the
Cantenna ever since, and still works well. *I do not use it much, but I
sometimes run 100w or about 700w (depending on whether the AL-80B is being
used) without problems. SWR indicated in the 1.1 or 1.2 range. *It was a
great investment.

Bill
W2WO

If you run packet on a radio connected to a cantenna, does that make
the cantenna a bit bucket?

Jimmie

JIMMIE wrote:
On Jan 24, 1:36 pm, "Bill Ogden" wrote:
I purchased a Cantenna from Heathkit in about 1961 or 62. I remember
finding a gallon of mineral oil in a drug store at a low price. (It must
have been a really low price to impress me at the time.) It has been in the
Cantenna ever since, and still works well. I do not use it much, but I
sometimes run 100w or about 700w (depending on whether the AL-80B is being
used) without problems. SWR indicated in the 1.1 or 1.2 range. It was a
great investment.

Bill
W2WO

If you run packet on a radio connected to a cantenna, does that make
the cantenna a bit bucket?

Jimmie

:-)