I am building an exciter for my 8000 rig. 6AG7 VFO, 6CL6 buffer and 807
final.

I am plagued with the problem that resonance in the final does not agree
with peak power out.

The power peak (watt meter or light bulb) is a considerable twist south
of the plate current dip.

Looking at the output on a scope shows nothing strange about the wave
shape as the amplitude increases beyond the dip.

I have tried neutralizing with coupled links between the input and
output tanks but it makes no difference nor does anything else I have tried.

Any ideas would be welcome.

Jack K9ACT


--
PHOTO OF THE WEEK: http://schmidling.com/pow.htm
Astronomy, Beer, Cheese, Fiber,Gems, Sausage,Silver http://schmidling.com

Hi Jack,

Have you tried removing drive to the PA and looked for output RF ?

Grid stoppers in place ?

Close, appropriate tight de-coupling to chassis (G2, A1 supply past
anode RFC) etc. ?

APC in the PA anode ?

Any chance of a photo to look at construction techniques ?

Regards,

Mark


On May 23, 5:23 pm, Jack Schmidling wrote:
I am building an exciter for my 8000 rig. 6AG7 VFO, 6CL6 buffer and 807
final.

I am plagued with the problem that resonance in the final does not agree
with peak power out.

The power peak (watt meter or light bulb) is a considerable twist south
of the plate current dip.

Looking at the output on a scope shows nothing strange about the wave
shape as the amplitude increases beyond the dip.

I have tried neutralizing with coupled links between the input and
output tanks but it makes no difference nor does anything else I have tried.

Any ideas would be welcome.

Jack K9ACT

--
PHOTO OF THE WEEK:http://schmidling.com/pow.htm
Astronomy, Beer, Cheese, Fiber,Gems, Sausage,Silverhttp://schmidling.com

Hi Jack,

Have you tried the PA without any excitation ?

Do you have an APC in the anode line, and an approprate RFC to HT
(with good decoupling to chassis) ?

How about grid stoppers ?

Regards,

Mark

On May 23, 5:23 pm, Jack Schmidling wrote:
I am building an exciter for my 8000 rig. 6AG7 VFO, 6CL6 buffer and 807
final.

I am plagued with the problem that resonance in the final does not agree
with peak power out.

The power peak (watt meter or light bulb) is a considerable twist south
of the plate current dip.

Looking at the output on a scope shows nothing strange about the wave
shape as the amplitude increases beyond the dip.

I have tried neutralizing with coupled links between the input and
output tanks but it makes no difference nor does anything else I have tried.

Any ideas would be welcome.

Jack K9ACT

--
PHOTO OF THE WEEK:http://schmidling.com/pow.htm
Astronomy, Beer, Cheese, Fiber,Gems, Sausage,Silverhttp://schmidling.com

On May 23, 12:23 am, Jack Schmidling wrote:
I am building an exciter for my 8000 rig. 6AG7 VFO, 6CL6 buffer and 807
final.

I am plagued with the problem that resonance in the final does not agree
with peak power out.

The power peak (watt meter or light bulb) is a considerable twist south
of the plate current dip.

Looking at the output on a scope shows nothing strange about the wave
shape as the amplitude increases beyond the dip.

I have tried neutralizing with coupled links between the input and
output tanks but it makes no difference nor does anything else I have tried.

Any ideas would be welcome.

Jack K9ACT

--
PHOTO OF THE WEEK:http://schmidling.com/pow.htm
Astronomy, Beer, Cheese, Fiber,Gems, Sausage,Silverhttp://schmidling.com

Jack, am a believer in the grid dipper,double check the buffer and PA
tuning ,make sure that they are on frequency and not doubling .If 807
has Pi network,make sure cap C1 nd C2 are in ball park for tuning
range of each band..Handbook wll give these values...Unplug 807 and
make sure the 6AG7 is on freq and on resonance.. think you hae a good
choice with tube lineup.just need to get it to settle down...GL 73
Harold W4PQW

On Tue, 22 May 2007, Jack Schmidling wrote:

I am building an exciter for my 8000 rig. 6AG7 VFO, 6CL6 buffer and 807
final.

It would help to get more details from you. What frequency is all of this
for? Link-tank output? Or pi-network? Are the coils of the tank/pi-net
according to specification (from some plans)? Do you know that the VFO is
on frequency? Are you doubling in the 6CL6?

I am plagued with the problem that resonance in the final does not agree with
peak power out.

Can you removed B+ from the VFO and twist your driver and final tune knobs
and get dips in plate, grid current (signs of self-oscillation)? Do you
generate grid bias in the final from a grid power supply or rectified
drive power?

I have had some experience with matching networks between the final and
the load having slight disagreements between peak output power and maximum
dip in plate current. Not to worry if the difference is slight.

Most important: find out if the final is self-oscillating (remove drive
and rotate final tuning knob to see if there is any wiggle in plate
current at any point).

The power peak (watt meter or light bulb) is a considerable twist south of
the plate current dip.

What is a considerable twist? five degrees of arc? or 45 degrees of arc?
What is your load? Dummy load?

Looking at the output on a scope shows nothing strange about the wave shape
as the amplitude increases beyond the dip.

All sine waves should look like sine waves.

I have tried neutralizing with coupled links between the input and output
tanks but it makes no difference nor does anything else I have tried.

This can be very difficult to do. If your frequency is HF or low HF you
shouldn't have to do this. Most old commercial HF tube rigs did not have
neutralization and they worked fine.

Any ideas would be welcome.

You do have the screen grids grounded through good RF bypass capacitors,
right? Once I forgot to add these to a 6146 amp and it self-oscillated at
160 meters (it should not do that) until I put in a good screen to ground
bypass capacitor.

Another area where you have to look out for a mistake is if you have the
coil/inductor at the grid input of the final located where it can pick up
the magnetic field from the final plate output coil/inductor, which can
contribute to making the final just a self-excited oscillator.

So, the next question is: if you get a dip in the final plate current at
some position of the plate tuning capacitor, then does that final plate
current change if you also tune the grid input capacitor? If yes, then
your final has a configuration to make itself a self-excited oscillator,
and the next thing is to look for how feedback can be contributing to
that.

Jack K9ACT


--
PHOTO OF THE WEEK: http://schmidling.com/pow.htm
Astronomy, Beer, Cheese, Fiber,Gems, Sausage,Silver http://schmidling.com

MarkAren wrote:
Hi Jack,

Have you tried removing drive to the PA and looked for output RF ?

Roger. Just pegs the meter with no output.

Grid stoppers in place ?

What's a grid stopper?

APC in the PA anode ?

What's APC?

Any chance of a photo to look at construction techniques ?

Sure. http://schmidling.com/807a.jpg

Needs a little cleaning up. Note the careful placement of the grid coil
at right angles to the final and with the Masonite panel between for
good shielding.

js

--
PHOTO OF THE WEEK: http://schmidling.com/pow.htm
Astronomy, Beer, Cheese, Fiber,Gems, Sausage,Silver http://schmidling.com

On May 23, 11:23 pm, Jack Schmidling wrote:
MarkAren wrote:
Hi Jack,

Have you tried removing drive to the PA and looked for output RF ?

Roger. Just pegs the meter with no output.

Grid stoppers in place ?

What's a grid stopper?

APC in the PA anode ?

What's APC?



Any chance of a photo to look at construction techniques ?

Sure. http://schmidling.com/807a.jpg

Needs a little cleaning up. Note the careful placement of the grid coil
at right angles to the final and with the Masonite panel between for
good shielding.


ERK! - looks like a 6146 PA, rather than an 807. And, congrats at the
attempt, (LOVE the agricultural look - wheres the PTO for the chaff
cutter?) BUT theres too much loose wiring providing stray coupling
paths - AND you do need to use a metal chassis, even if only to
separate the input/output circuitry. A piece of light gauge aluminium
sandwiched(sp?) on top of your present chassis would do it.....easy to
work with simple tools, no bends.....or do what the old timers did,
go to the supermarket, buy a sheetmetal baking tray from the Homewares
section)

BTW - while your there, buy one of those cheap white plastic kitchen
cutting boards - easy to work, a lovely source of insulating material
for standoffs, insulators etc...

And whats with the 3 big wire wound resistors? - screen supply to the
PA - what a potential source of stray coupling. A 1 watt resistor is
usually sufficient, you wont need the string of 10 watt ones....
(especially sitting under the lead from the anode of your PA stage..)
And the anode tuning is about 100 yards from the tube! - can you get
it any closer?

So saying all that, general rule of RF stability is to bypass
EVERYTHING, including any existing bypass circuitry. (This is why a
metal chassis is so good...no long leads on bypass caps)In this HF
cct, a handful of 0.1 or even 0.01 caps would be a good try....wack
em everywhere. Go buy heaps, they will be most useful for other stuff
as well. 630V polyester/polystyrene are easily and commonly and
cheaply available...

And keep going - its worth the effort -

Andrew VK3BFA.

Hi Andrew,

Yes, the example I quoted was for a 6146, but same construction and
circuit elements apply.

I just wanted to give concrete examples of the elements I was talking
about.

-M

On May 25, 1:17 am, Andrew VK3BFA wrote:
On May 23, 11:23 pm, Jack Schmidling wrote:

MarkAren wrote:
Hi Jack,

Have you tried removing drive to the PA and looked for output RF ?

Roger. Just pegs the meter with no output.

Grid stoppers in place ?

What's a grid stopper?

APC in the PA anode ?

What's APC?

Any chance of a photo to look at construction techniques ?

Sure. http://schmidling.com/807a.jpg

Needs a little cleaning up. Note the careful placement of the grid coil
at right angles to the final and with the Masonite panel between for
good shielding.

ERK! - looks like a 6146 PA, rather than an 807. And, congrats at the
attempt, (LOVE the agricultural look - wheres the PTO for the chaff
cutter?) BUT theres too much loose wiring providing stray coupling
paths - AND you do need to use a metal chassis, even if only to
separate the input/output circuitry. A piece of light gauge aluminium
sandwiched(sp?) on top of your present chassis would do it.....easy to
work with simple tools, no bends.....or do what the old timers did,
go to the supermarket, buy a sheetmetal baking tray from the Homewares
section)

BTW - while your there, buy one of those cheap white plastic kitchen
cutting boards - easy to work, a lovely source of insulating material
for standoffs, insulators etc...

And whats with the 3 big wire wound resistors? - screen supply to the
PA - what a potential source of stray coupling. A 1 watt resistor is
usually sufficient, you wont need the string of 10 watt ones....
(especially sitting under the lead from the anode of your PA stage..)
And the anode tuning is about 100 yards from the tube! - can you get
it any closer?

So saying all that, general rule of RF stability is to bypass
EVERYTHING, including any existing bypass circuitry. (This is why a
metal chassis is so good...no long leads on bypass caps)In this HF
cct, a handful of 0.1 or even 0.01 caps would be a good try....wack
em everywhere. Go buy heaps, they will be most useful for other stuff
as well. 630V polyester/polystyrene are easily and commonly and
cheaply available...

And keep going - its worth the effort -

Andrew VK3BFA.- Hide quoted text -

- Show quoted text -

Andrew VK3BFA wrote:

Needs a little cleaning up. Note the careful placement of the grid coil
at right angles to the final and with the Masonite panel between for
good shielding.

ERK! - looks like a 6146 PA, rather than an 807. And, congrats at the
attempt

I thought the "Masonite shielding" would suffice for a smiley.

That was my first attempt at a wooden chassis and it became obvious why
metal is better and I abandoned it a few weeks ago.

It started out xtal controlled but then I got the bug to add a VFO which
is the crap on the left. The 807 turned into an 866 so I substituted
the 6146 till I got another 807. The missing meter was a hamfest
special that bounced so much it was useless and then crapped out completely.

It is now all neatly mounted on a home made aluminum chassis and
unfortunately looks a lot better than it works.

I will take a for real pic tomorrow. That one was just too good not to
share.

And whats with the 3 big wire wound resistors? - screen supply to the
PA - what a potential source of stray coupling. A 1 watt resistor is
usually sufficient, you wont need the string of 10 watt ones....

One of the plans I was working for called for 25W. Seems like a
misprint because it does not even get warm. The 10K in the VR150 leg
gets hotter than hell and probably should be 25W. I burned my finger
just checking it.

And the anode tuning is about 100 yards from the tube! - can you get
it any closer?

If you look closely, you can see the 807 socket between the 6146 and the
cap. The 6146 is where the other 6ag7 is supposed to be.

And keep going - its worth the effort -

It's been a challenge. I made a simpler version of this as a kid and am
sort of re-living my childhood. It's like a Mr Wizzard Science project
for me.

I am now plagued with real test equipment and can't play "ignorance is
bliss".

js

--
PHOTO OF THE WEEK: http://schmidling.com/pow.htm
Astronomy, Beer, Cheese, Fiber,Gems, Sausage,Silver http://schmidling.com

On May 25, 2:31 pm, Jack Schmidling wrote:

I thought the "Masonite shielding" would suffice for a smiley.

It did - I just didn't want to spend the next 3 weeks figuring out how
to put one in.....(bloody computers)

And thanks for the explanation - I didn't see the 807 socket, and
didn't know there was a VR150(was it?) in there as well. And the
extraneous tuning stuff to the left, well, now I know its for a VFO.
(Gee, voltage regulation - thats getting fancy - yes, the dropper
resistors do run stinking hot, don't they....)

Strangely enough, theres a fault in human design in that the nerve
endings that detect heat are a lot bloody slower than everything else
- by the time we figure out its too hot, the damage is done....

And must admit to wallowing in nostalgia - mine attempts were
similar, and probably oscillated all over the place as well - BUT
without virtually any test gear, (or knowledge) it didn't seem to
matter - there was no one to complain anyway, the bands were virtually
empty except for other manic experimenters, TV was black and white,
and broadcast was AM. Nothing else to get into - and the bands were
harmonically related so we we're QRM other hams anyway....2m was the
outer frontier anyway, there was NOTHING used above that....

How about a photo of the latest metal attempt? - remember, if it looks
nice and neat, it doesn't really matter if it don't work too good....

I have things here I am not game to connect to the spectrum analyser -
don't want my dreams shattered toooo quickly....
but getting back, more and more, into the older stuff - partly fading
eyesight (I cant see u-mini smd ic's) and besides, theres something
comforting about the warm glow of valves in a dark room, the crash of
static from the other side of the world, the thrill of hearing someone
on air from an exotic National Geographic land...the reasons we got
into amateur radio in the first place..

The bands here at the moment are a total write off everywhere - 80m is
S9 noise, the rest not much better. QRM, QRN wall to wall. So, I sit
here and type on a keyboard, I guess its still "hamming" after a
fashion...



That was my first attempt at a wooden chassis and it became obvious why
metal is better and I abandoned it a few weeks ago.

Yep. Tubes with gain killed wooden chassis.....

It started out xtal controlled but then I got the bug to add a VFO which
is the crap on the left. The 807 turned into an 866 so I substituted
the 6146 till I got another 807. The missing meter was a hamfest
special that bounced so much it was useless and then crapped out completely.

Hah! - silly man - NOW you have to get the VFO stable, as well as all
the other problems. And as for 807's - I think about 90% of the people
reading this will have some sitting around, just waiting for the
"magic moment" of inspiration. Good On Ya Jack - your doing it - good
luck!

And the magics still there, even after 30 plus years of
experimenting........cant be killed by the internet, store bought
gear, mobile phones - all those glittering toys that didn't even
exist exist when we started out.....

And I got boxes of hamfest meters that are probably crappy too -
haven't tried to use them yet, so don't know. Finally built myself a
rough tube tester a few weeks ago (I had to, trying to find working
tubes for a VERY elderly Yaesu FTDX400)


It's been a challenge. I made a simpler version of this as a kid and am
sort of re-living my childhood. It's like a Mr Wizzard Science project
for me.

I am now plagued with real test equipment and can't play "ignorance is
bliss".

Yep. Its a real PITA that - we cant delude ourselves our home brew
QRM generators are not what they are.....still, its an excuse to use
some time creatively.......



73 de VK3BFA Andrew.