[AMRadio] RE: CE 20A Phase shift question

James M. Walker chejmw at acsu.buffalo.edu
Thu Feb 2 22:32:30 EST 2006

Hi All,
I am going to answer the comments but first remove some of the

> My VFO458 units are stock, as in just like the manual, with changes
> says, and I have not experienced this phenomenon at all!
I found MOST of the changes to the Central Electronics version of the
VFO458 by buying more than one, I currently own three, which came
with manuals and most of the manuals, had change notices in them. From
those after putting them in sequence, I made the changes necessary to
eliminate most of the problems folks say they have with the VFOs.

Such as the heat generation problem was solved by placing the dropping
resistor on the outside of the unit. The hum you speak about (60 hz) was
cured with two changes, one was using DC on the filaments of the 1625.
The other was to match the hot sides of the two chassis VFO and exciter
either 10A, 10B or 20A and eliminate the voltage difference between them.
Hum was down, no problems when doing balance adjustments on the exciters.

Yes as I mentioned my VFOs, are stable, to the point that after about five
minutes of warmup the frequency is right where I left it. I check my freq.
output with a high standard HP counter, and a Tek spectrum analyzer.
I have not had any complaints about frequency stability while on the air.

> Carrier null on several of my units is in the -80 db range, and stays that
> way, while in use.
> WD5JKO Replies: With those carbon 1 turn null pots, a deep null is 
> possible,
> but hard to keep there due to temperature sensitive components in the
> balanced modulator, and the physical issue with the coarseness of the 
> carbon
> pots themselves. The null is limited by hum modulation both 60 and 120 hz.
> The 120 hz is easy to fix with an additional RC filter feeding B+ to the 
> two
> modulator tubes. The 60 hz component is from stray magnetic fields in the
> chassis coming from the power transformer. I did reduce it in mine by
> injecting 60 hz in the correct phase and amplitude into the chassis to in
> effect "neutralize" that hum as seen in the nulled carrier. The point of
> injection is critical, and was best placed at a specific spot on the
> balanced modulator diode mounting bracket. My null is at best -60 dbc, and
> yours as claimed is as good or better than state of the art equipment
> manufactured today.

When doing the refurbishing, in some cases I installed new pots, where they
were intermittent, but mostly it was not needed. I did not notice stray 
around the power transformer area, just signs of over heating, a changing
of some power resistors, and a few electrolytic caps cleaned up a lot of
the hum and intermittent noise on the signal paths. Also good 6U8 tubes
were put in the units that use them, as the mixing function was erratic at
State of the art suppression can be achieved with test equipment and
patience, also good components are a real necessary part of the process.
The 60 hz and 120 hz components are covered above. As for State of
the Art, that is determined by how you do the setup, the equipment you
use in the testing and evaluation., and proficency using it.

> My CE gear is all stock with refurbishing done to them to bring to
> manual specs and ranges. The 10 meter output on them are at 37.0
> to 38.7  Mhz.
> I have run them on 10 driving a Johnson Thunderbolt, without any
> drifting or squirrelly effects, and great power out right around 300
> watts carrier to the antenna.

Maximum output with the refurbished 20As is 10 watts out on 160M dropping
to 5 on 10M. My 10Bs can only do 5 watts on 160M and 2.5 +- on 10M.
All of my VFO458 units are setup to cover 160M to 10M per the conversion
manual and changes thereto. The trick is to get 37 Mhz output that is stable
and clean. Takes some fussing around but it can be done.

> WD5JKO Replies: What power does your 20a provide on 10 meters? I assume 
> that
> your VFO has the 10m option in it.
> The manual calls for 115 VAC to the power transformer, at
> that input voltage, and with good tubes, 5U4 and 6AL5 the
> voltage levels are pretty close to the manual, 310VDC output
> in my case. The 6AG7s normally run hot in "AB(*) operation.
> CE put out change notices to lower the screen voltage to the
> proper level and also to move the paracitic suppressors to the
> output of the tubes and not the input.

> WD5JKO Replies: Yes running these on 115 volts is easier on this rig, and
> most boat anchor gear as well. Still, and stock 20a plugged into 125 volts
> will put out about 20 watts (160, 80m), maybe 15 on 20m, and about 8 watts
> on 40 meters (this is where the VFO triples). They run hot that way, and a
> fan is required. I took the approach of undergoing a QRO project for the 
> 20a
> using different tube(s), and higher plate voltage, but that is another 
> story
> that I've covered here before.
I have not been able to ever get 20 watts output from the 20As, but then I 
run them from a variac set at 115 VAC. However I do use the recommended
RCA 6AG7 tubes and no others in the output section. The tubes need to be 
and again the 6U8 is the main thing, the mixer tube MUST be up to it or the
drive to the succeeding stages will fall off drastically, causing lower 
Also if the changes to the grid input of the 6AG7 tubes is done the output
shows a greater and cleaner output.

For use with the ThunderBolt amp, the drive needs to not exceed 5 watts no 
which band you use. On 80M I generally run the 20A at 5 watts and
tune the Tbolt for 400 Watts of output into the dummy load, I monitor
the waveform out and make sure the grid and screen currents do not go
off the recommended levels.

> My 20As run right around 7.5 watts output on AM, and 12.5
> on CW. If you are going to drive a linear amp like the Thunderbolt
> you really need to monitor the power output from the 20A to the
> Tbolt, and keep it around 2.5 watts.
> WD5JKO Replies: Yes in a stock 20a on 80, 160m, there is plenty of drive
> available, and the transition from class Ab1 to Ab2 is at about 12 watts. 
> In
> Ab2, the bias source is high impedance, so self limiting occurs, and this 
> is
> seen by gain compression between 12 watts and maximum power. Therefore for
> AM, 115 vac operation, 3 watts AM (12 watts PEP) works out real nice, and
> perfectly matches amplifiers like the T-bolt, and CE 600L amplifiers. At 5
> watts AM, a 20a looks like hell on the scope due to gain compression on 
> the
> positive modulation. I don't understand how you can achieve 7.5 watts AM 
> on
> your 20a when the max cw is 12.5 watts unless you are limiting your 
> positive
> modulation to something under 50%

> The rigs were never designed to provide 20 - 25 watts, their main
> intent was to be a low cost low power exciter with clean output, to
> drive a higher powered linear amplifier. Such as the Tbolt or the
> 600L pair of 813s.
> WD5JKO Replies: Amen
> Jim
> Hope this wasn't too confusing.
> WD5JKO Replies: No not confusing, but some of what you said sure seems to
> indicate that your 50 year old CE 20a is as good as more modern gear, or
> maybe your recollection is painting things in a better light.

State of the art again is defined by, the folks doing it. The signals should 
as clean as possible, consistant with good engineering practice. My 50 plus
year old gear, is as good as any of the current state of the art, if what I
hear on the bands these days is any indication. Even with totally digital
frequency readouts, most SSB signals are off frequency with the folks
they are talking to by as much as 1.5 Khz. Whereas I monitor where my
signals with equipment that tells me where I am transmitting, what my power
output is, and what both the transmitted and received (at my QTH) signals
look like. Radio IS, if it worked once it can be made to work again, just
depends on how you go about it. Of course "having a ferarri doesn't,
automatically confer Mario Andretti status on a person" so sometimes
the volkswagen wins!

More information about the AMRadio mailing list

This page last updated 22 Oct 2017.