Power Supplies


Jim, has your opinion of the MFJ power supply changed or do you
still recommend it over other switching power supplies? Anyone
else using it?


I'm gonna be ordering the MFJ-4245MV later today... This is the
45amp "Big Brother" of the MFJ-4225MV that Jim reported on.
Actually it's only a pound And a half heavier and slightly
bigger... I don't want to have to reduce voltage To 10v and use
lo power setting on the amp just to get the tube lit as Jim

I've been spoiled by my Astron 50 amp VS-50M.... 46 pounds of
solid, indestructable Power!....


> I'm gonna be ordering the MFJ-4245MV later today... This is the 45amp "Big
> Brother" of the MFJ-4225MV

I did not know there was a larger one. Good! I was wondering
if there was a larger one since Jim had problems with the 25A
version. Do you have an idea on cost? Do you have to order
from MFJ or is Left Coast carrying them?

> I've been spoiled by my Astron 50 amp VS-50M.... 46 pounds of solid,
> indestructable Power!....

That's the way I feel about even the 35M. Bulletproof. I
think it is because I use short cables, but I have never read
over 11A on the meter, so 25A would likely be fine, but I
would get the 45A for the extra margin anyway if it is
reasonably priced.


I have purchased no less than 4 different switching power
supplies and found only one that ever worked correctly and
wasn't subject to a feedback "hash". This hash sound would
interfere with the operation of the power supply and upset the
devices wave. I have discovered that the way to fix this is to
add more capacitance across the output terminals of the power

I got a 4700 uf electrolytic cap from Radio Shack, and
connected it across the output terminals of the power supply.
Negative end to the negative output, and positive end of the
cap to the positive output of the power supply. This pretty
well solved the hash problem and made the system sing very
nicely .

A bit more capacitance should optimize this fix. I would say
look for about a 6000 uf (or mfd - same thing ) cap at 20 volts
minimum to bridge between the output posts of the power supply.

This applies to switching power supplies only! These supplies
are the PC30 which a lot of people have purchased, the Kenwood,
Astron, and the Samlex.

The arrows on the side of the capacitor point to the negative
end! Do not install backwards, the cap will literally explode!



Another though here, if hash is a problem is to use a car hi fi noise
suppressor. These things usually consist of one or two inductors and a
capacitor or two for much greater attenuation of garbage.

Trevor Wilson


The MFJ 45A switching power supply (non hashing) is
now available from MFJ for $199, or the following:

Texas Towers, 1108 Summit Ave., Ste. #4,
Plano, TX 75074
(972) 422-7306 / (800) 272-3467


> I'm researching hardware, and am wallowing in freedom of choices.
> I'm inclined to choose the Pyramid PS-52K over the Astron RS-50A or M --
> -- (based primarily on cost/power tradeoffs and some other individual
> recommendations.)
> What's your opinion? quality? ...

I don't have experience with the PS52K, but do with their 30A.
Until recently, people had problems with Pyramid supplies due
to RF noise, but they recently started building RF filters into
them, and they work fine - something to consider if you are
looking at used ones.

The MFJ45 works fine, too, and is also a switching power supply
with good filtering. Considering it is 45A and about $200 and
has lighted gauges, it is probably a better deal than the PS30
for $180. But, my current portable unit has a PS30 and works
great. Looks good too since it is covered with cooling fins,
plus has a thermostat controlled fan, but no ON-OFF switch
(which is good to have when there's a Kinnaman and you want to
keep the fan noise down when programming it.)

> (FYI: I decided on a larger supply than the RS-35A in Dr. Bare's book
> to allow for growth and expansion.)

Unless you are planning on using a 500W or so amp, I could not
imagine needing a power supply larger than 35A. The least
efficient setups (long cables, crummy tube wrapping => hot
components) I have tried with max power on the amp have not
exceeded 15A, and I typically run about 10-12A at max power on
the amp with good setups.

> Is the 52K acceptable? (as per harmonic suppression?)

You don't need to worry about harmonic suppression with power
supplies, they just need good noise filtering. I still put
clamshell or torroid chokes on all inputs and output of the
power supplies I use.

> Can it output a clean lower power output like the Astron RS-35A?

I have never used it, but I imagine its RF filtering works as
well as on the 35A. BTW, the 35M is only a few dollars more
and gives you voltage and current gauges.

You do know that the difference between the, say, MFJ45A and
the Astron 52? The MFJ weighs about (guessing here) 6 pounds
and the Astron about 40. Definitely something to consider if
you ever plan on moving the device. Back in the days when
Astron was the only recommended supply to use, I had a very
strong (and heavy in itself) rack to hold the BRG, and had to
make the power supply easy to remove since the combination was
so heavy.

> Problems? Horror stories?

Just about the hernia-inducing lugging of my BRG when it had an
Astron in it.

I still think the Astron is superior to switching power
supplies, and would use it if portablity was not an issue.


Tech America once again has the Samlex 1223 switching power supply on sale.
This is a dual volt 110-120 / 220-240 unit. The same unit is used by Don
Tunney in his units. 13.8 volts @ 23 amps. For many of our applications, it
will need the addition of about 10,000 to 20,000 uf @ 16 volts or more
electrolytic cap across the output terminals .

Price on sale is $79.95 each Shipping and handling charges vary but are
very reasonable

Item # 910-3916 www.techam.com

24 hour fax 1-800-813-0087
24 hour phone 1-800-877-0072


Recently, we encountered a problem with our 1223
switching power supply. We had been running about 15 minutes,
driving a small linear amplifier which was supporting about a
50 watt resistive/capacitive load.

The input program switched from around 800hz to 20hz,
whereupon the controlling software placed the system in PAUSE.
Proper adjustments for low-frequency inputs were executed
and.....the instant the system input signal was
reactivated----BOOM, there goes one of the big switches, plus a
couple of peripheral components.

V.T. Singh, one of the lab supervisors at Samlex (in
Canada) called me, upon their receipt of our failed device and
related the following: First, this power supply has NO
high-capacitance power filter directly in front of the
switches; nor does it have a fast-rise clamping circuit on the
output (although it does possess normal current limiting in the
output). Mr. Singh informed that due to the nature of the
linear AMP reflecting ANY sizeable, instantaneous, load
increase directly to the power supply......at high
speed....places the power supply in the position of looking at
EXTREMELY steep current demand spikes. Essentially, you have
an avalanching condition taking out the switch in a VERY SHORT
TIME! An this, no matter if the AVERAGE LOAD is well within the
nominal operating current limits. The problem seems to be that
the switches are wiped out INSTANTLY....before the current
limiting function can respond to this, particular, state of

After considerable, confusing, exchange, I gleaned that
short of straping a fully-charged automotive battery across the
output of the 1223, there is no guarantee one is not subject to
random repetition of this failure.

Mr. Singh related a very interesting observation to
me: "Switching power supplies, of this type, DO NOT LIKE LOW
FREQUENCY AUDIO LOADS'. (Caps are my emphasis).

Although he did not overtly recommend such an addition,
Mr. Singh did appear to, passively-agree to my suggestion that
1500-2000 mfd of electrolytic across the output of the 1223
would mitigate against most such occurrences.

I would be interested to know if anyone else in our
little community has experienced any kindred behavior in their
SEC 1223?

Sincerely, Brooks Bradley.

P.S. The Samlex people were very cordial and accommodating and
are repairing the device at no charge. Their people appear
very intelligent and quite technologically competent.


saw the problems people had with their power supplies I thought
you might of put what I did on it... adding lots of capactance
to it...since it made a difference in how the beam felt to me.
I see a few thousand microfards added, is suppose to make a
difference on your webpage... really I have found that MORE is
better..it just feels better. Probably the one one that it
might not of made a difference on is the Astron like the one
you and I both have..I added more to it anyway.. WHY? because I
could and there was plenty of room to spare inside of the
case...I will eventually bring it up to 100,000 mfd...just
because I can...and it makes me feel better knowing that I at
least did something.. hey I cant leave well enough alone..ha
ha...you know that, Brian..I GOTTA tinker with it some..ha ha
on a large scale assembly line they cant do that..because of
the cost....on a 1 on 1 basis, you can do that...modify it....

it really does make a difference in how it feels and really
helps it operate better.

caps are cheap...your try it yourself and see...okay yeah I
know, you dont feel the beam...so maybe someone else you know
who can feel and has had alot of exposure to it, might be able
to feel the difference... anyway its a cheap modification to
do..and people buy what they can afford... yeah sure I want a
super clean power supply thats light weight and its out of my
budget...so I make do with what I can and beef it up...works
for me and it will work for you too, okay


I found a source for some really big caps in the correct
voltage - Cheap!

Try Hosfelt Electronics 1-800-524-6464 phone
Fax 1-800-524-5414

Out of US phone 740-264-6464

Hosfelt Electronics
2700 Sunset Blvd.
Steubenville, OH 43952-1158

89,000 mfd @ 15 volts item # 15-835 $5.95

55,000 mfd @ 30 volts item 15-761 $9.95

Others are available at similar prices

Warning!!! Large caps such as these will store a very large
charge of electricity. They can give you a very nasty shock
even when the power is off !

Jim Bare

>So how much extra added captance will I add?..A WHOLE BUNCH...
>My PC-30 didnt start to feel smooth till I added 45,000 mfd to it...it
>now has more, because that is only what I could feel..
>This Samlex...well...no sense in making several trips to the electronics
>store...I will probably add A WHOLE BUNCH to it also...about 80,000 mfd


Well over a two years ago I recieved a couple of reports from
people that using two separate power supplies on the device
seemed to help the effects. The recommended set up was to use
one small supply for the CB and a large supply for the linear.

I tried this and although there was some minor enhancement, I
never said much about it, as I didn't really think it worth the
expense or effort. In fact I have continued to use just one
power supply for my personal systems.

For whatever reason I decided to take another look at using two
power supplies this last week. I used one for the function gen
and CB, and one solely dedicated to the linear. Not much
difference with standard continuous output, but... when put in
pulse mode the difference in the pulsing tube was quite
significant. In fact the difference was enough that I will be
picking up some small power supplies to run my CB at least
separately from the linear.

I think what is happening is that the pulse mode upsets the
supply, it sees a small load for the CB and function gen
first.Then a few microseconds later the linear puts load on the
power supply. At a certain gate rate the system shuts down and
then restarts again. Puts a lot of stress on the power supply
doing this. Result is a compromize from the power supply, it
just does the best it can. A small 4 to 5 amp power supply is
all that is needed for the CB, function gen and the lights .

The extra power supply enhances the device only slightly in
continuous mode as there is a rhythym and timing to the pulse
that gives the power supply time to adapt.

Worth trying, and highly recommended if you tend to use the
pulse mode on your frequency generator.