Rich Measures wrote:
>>In message <PKUw4HABVmmzEwFZ@ifwtech.demon.co.uk>, "Ian White, G3SEK"
>><G3SEK@ifwtech.demon.co.uk> writes
>>>...snip......... Also
>>>the 91b has a low-impedance grid bias supply, so small amounts of grid
>>>current do not spell IMD "nutsville".
>>
>>I'm learning a lot here, ...snip,,,
>
>Yes, but does it wash? When a tube with a high output-R (100k ohm) bias
>supply is temporarily overdriven by a voice peak, the negative grid bias
>voltage temporarily increases due to grid-rectification and the resulting
>E=I*R drop across the 100k resistance, so the grid becomes more negative,
... which causes gain compression and IMD.
>so grid current tends to decrease, which limits IMD-genenrating grid
>current.
That's a circular argument. You're ASSUMING that grid current causes
IMD, so of course you must "deduce" that anything that limits grid
current must reduce IMD.
>OTOH, with a stiff grid bias supply, overdrive produces high
>grid current. ....... Does increasing grid current in Class AB1 increase
>or decrease IMD?
We all know that gross overdrive will cause IMD, but we're talking here
about small amounts of grid current (eg enough to actuate the ALC).
With small amounts of grid current, some IMD products will go up but
others may come down. It all depends on the magnitudes of the 3rd, 5th,
7th and higher-order terms in the Vg-Ia transfer function, which in turn
will depend on the type of tube and the DC operating conditions.
Linearity of the transfer function will also depend on how the DC
operating conditions change at the onset of grid current, and whether
the grid swamping is sufficient to prevent a major change in RF input
impedance at that point in the cycle where the tube itself starts to
draw some drive power.
One thing is certain, though: anything that puts a major kink in the
transfer characteristic will increase higher-order IMD. The higher the
output resistance of the grid supply, the stronger that kink will be
where the instantaneous grid voltage crosses zero.
A "read-only" member of this mailing-list has reported an improvement in
the IMD of a pair of 4-250s on going from class AB1 to class AB2. This
may be surprising, but knowing his identity I trust his data. It is no
coincidence that he uses a shunt-stabilized grid bias supply with a low
output impedance.
Crossing with this download is a message quoting some very impressive
Eimac IMD figures from the 4CX1500B running in class AB2 with a small
amount of grid current.
I'm not saying that the IMD gets any better with a small amount of grid
current (enough to generate ALC). What I am saying is that with a low-
impedance grid bias supply, and also effective swamping of the impedance
change, IMD doesn't have to get dramatically worse.
73 from Ian G3SEK Editor, 'The VHF/UHF DX Book'
'In Practice' columnist for RadCom (RSGB)
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|