[Amps] a common topic here tspa

[R.Measures]

On Apr 5, 2005, at 1:00 PM, Joe Subich, K4IK wrote:

>
> Rich writes:
>
>> Joe -- Are you suggesting that Eimac's 8877 development team grounded
>> the grid through long thin wires?
>
> As I recall, and it has more than 20 years since I was able to set down
> with some of the best tube guys at Eimac over beers

Of course!

> (while we were taming
> "experimental" UHF TV transmitters).  Since all of us were hams, the
> conversation always turned to amateur amplifiers ...
>
> They told me that the grid resonance in the 8877 was above the plate
> resonance

The "plate" (anode) resonance is not fixed, instead it varies according 
to the total lead inductance between the anode (10pF for the 8877) and 
the Tune-C -- so, no tube  has a "plate" resonance per se.   Thus, each 
layout has such a resonance.

> and if the grid were properly grounded (a low inductance,
> low impedance connection), the phase relationship between the grid
> and anode was such that the tube would not oscillate.

During the grate parasitics debate with your friend Tom Rauch, Jr. I 
received a telephone call from a Ham who worked in the design 
department of a amplifier mfg.  One of the computer programs they used 
allowed the designer to foresee stability problems by entering 
feedback-C, input-C, input-Z, and gain-frequency data.  He said that an 
8877 was capable of oscillating above about 110MHz.

> The "problem"
> with the tube was those designs that used "long" thin leads from the
> grid pins on a socket ... and worse yet were those that inserted
> other devices and attempted to bypass them with capacitors.

Joe -- Are you saying that Eimac's 8877 design team told you they did 
not choose to use Eimac's grid-grounding collet during testing of the 
8877?
>
> Their advice was to use a collet of finger stock, or at least the
> Eimac socket with four wide grid contacts bolted directly to the
> chassis and do any grid current metering, including a grid current
> trip, in the cathode (float the B-).
>
> 73,
>
>    ... Joe, K4IK
>
>
>
>

Richard L. Measures, AG6K, 805.386.3734.  www.somis.org