[Amps] 8877 Max Grid Current

[Richard]

Trivia questions' solutions:
The minimum anode-to-grid potential for an 8877 is c. 250V.  Thus, with a 
4000Vdc anode supply, the anode voltage swing is c. 3750V-peak, or 
2650Vrms.  At 250 MHz the reactance of the 10pF anode-grid C is 
63.7-ohms.  Since I=E/R, the AC grid-current = 2650Vrms/63.7-ohms = 
41.6Arms.
At 144MHz, the Xc of 10pF is 110-ohms.  The AC grid current is 
2650Vrms/110-ohms = 24Arms.
-  In both cases, the DC grid current is probably c. 100mA.
As can be seen in Figure 24 on my Web site, the 8877 grid cage is made of 
over 100 bars of rectangular metal [gold-plated molybdneum].  Thus, at 
its maximum ratings, each grid bar is carrying only 0.38Arms plus 9mAdc.
-  notes:  the 8877's grid should be grounded with the requsite collet to 
reduce L.  The best protection against potentially-fatal heater/cathode 
arcs is to connect the heater to the cathode and feed the heater through 
a 10A bifilar choke.  Also, neither side of the heater should be grounded 
And the HV-negative/cathode-return should be protected with glitch diodes 
to limit voltage rise during a +HV to ground arc.  

have fun
-----------------
>
>
>>Rich,
>>
>>My data sheet shows 25 watts which works out to be a grid current of around
>>380mA.  This seems too high.  I fear that it may not be that simple (mixing
>>AC and DC characteristics) although my tube theory is fluffy at the best of
>>times.
>>
>>Popular opinion from all who have responded to me (and thanks to you all) is
>>a maximum of between 80 and 100mA.
>
>The 8877 grid is widely viewed as extremely delicate.  My guess is that 
>this is due to the tendency of gold to evaporate from the grid and cause 
>sometimes-fatal grid/cathode arcing.  Roughly half of the 8877s that I've 
>autopsied became kaput from this phenomenon.  In Feb., 1986, a tube specs 
>guy at Eimac (W. Foote) told me that gold evaporation was the result of 
>vhf or uhf oscillation.  He described the autopsy process used at Eimac.  
>I decided to have a go at it.  I found that the gold had evaporated from 
>the grid, that the cathode was coated with numerous gold melt-balls.  
>There were also gold melt-balls on the anode insulator - which expained 
>the leakage current indication for the tube prior to autopsy.  The gold 
>evaporation pattern from the grid indicated that there had been more 
>current at the grounded end of the grid than at ungrounded end -- much 
>like a quarter-wave vertical antenna.  I concluded that the frequency of 
>the current that evaporated the gold was likely in the uhf region.  At 
>Figure 24 on my Web site, there is a photo of an autopsied 8877 that died 
>from gold evaporation.  [note -- gold evaporation does not occur until it 
>melts and boils.  At one-atmosphere of pressure, gold boils at 2966ºC].
>-  One computerized amp design program/app. predicts that the 8877 has 
>enough internal feedback-C plus enough high-frequency gain, to become 
>marginally stable above 100MHz.  One way to address this problem is to 
>artificially reduce high-frequency gain through the use of a vhf/uhf 
>parasitic suppressor. 
>>
>cheers, David
>
>Today's trivia questions:  How much AC grid current flows through the 
>grounded 8877 grid at it's max ratings of 4000Vdc and 250MHz?  How much 
>at 144MHz/4000Vdc?
>
>
>
>>VK3HZ
>>
>>----- Original Message -----
>>From: "Richard" <2@mail.vcnet.com>
>>To: "David Smith" <vk3hz@wia.org.au>; " AMPS" <amps@contesting.com>
>>Sent: Tuesday, June 18, 2002 11:57 AM
>>Subject: Re: [Amps] 8877 Max Grid Current
>>
>>
>>>
>>>
>>> >I'm currently building a 2m 8877 linear amp, based on the K1AGB design