Amps
[Top] [All Lists]

Re: [Amps] a common topic here tspa

To: "Joe Subich, K4IK" <k4ik@subich.com>
Subject: Re: [Amps] a common topic here tspa
From: Richard Measures <r@somis.org>
Date: Fri, 8 Apr 2005 18:30:09 -0700
List-post: <mailto:amps@contesting.com>
On Apr 8, 2005, at 2:39 PM, Joe Subich, K4IK wrote:

>
>> From: Richard Measures
>>
>> Denny Had did not use an Eimac grid-grounding collet, and that
>> increased the chance of a parasite.
>
> No fooling - and exactly what I've been saying.  However, that is
> not the only reasonable explanation for grid overheating, particularly
> in a tube known for running excessive grid current when over driven.

Over driving with even 200w could not heat the grid to the temp needed 
to sputter gold.   I have autopsied a number gold-sputtered 8877s that 
were driven with 100w radios.
>
>> Eimac told me that the 8875s I
>> sent them from a MLA-2500 were probably damaged by an oscillation
>> condition.
>
> Foote told you the tubes had been "badly overheated internally."
> He ascribed that overheating to an "oscillation condition" but
> you had already given him that "parasite" diagnosis.

Eimac was the first source of the oscillation explanation.
>
>> Are you suggesting that overdrive could melt and evaporate gold?
>
> I am saying that any mechanism that causes overheating of the grid
> can cause damage, including stripping (or evaporation) of the gold
> plating.

No matter how many watts?
>
>>>    4) the tube is not well cooled (same internals as the 8874 with
>>>       a less efficient anode cooler)
>>
>> 8874s suffer from gold sputtering.
>
> Of course, they are also very easily over driven.  If I recall the
> 8873/8874/8875 is rated for 20 to 25 watts peak (cathode) drive and
> a maximum grid dissipation of 5 watts!
>
>>>    5) the socket end of the tube gets almost no cooling air in
>>>       the MLA2500
>>
>> Taurine feces.
>
> DO you have real measurements?  If so make a rational argument -
> don't stoop top playground scatology.

The MLA's fan blows directly on the hottest part of the sockets.

> I've had an MLA2500 and
> have seen how hot the socket/chassis gets in comparison to other
> amplifiers with proper cooling.

You have extraordinary vision.
>
>>> Your claims that the presence of gold melt balls as proof of
>>> parasitic oscillation is quite simply a case of hearing hoof
>>> beats and looking for zebras.
>>
>> The idea wasn't mine, it was Eimac's 8877 development team's.
>
> The development team did not posit VHF parasites as you have been
> claiming.  Their explanation was a high UHF or SHF oscillation.

I said it was UHF.  Foote told me that the 8877 dev. team did not 
measure the frequency of the oscillation.

> The same problem befell another Eimac product with which I was
> involved ... it caused localized heating of the output window and
> resulted in sudden, premature, catastrophic failure.
>
>> I thought electrons travelled toward the anode, not toward
>> the cathode.
>
> The grid is still in the electron beam and intercepts some of the
> anode current.

The grid intercepts some of the cathode current, Joe.  The anode gets 
what remains

> That current (particularly when over driven) results
> in overheating of the grid.  Electrostatic attraction will assist
> in stripping the coating from an overheated grid.

So why do a good number of gold meltballs stick to the cathode?

>
>> The evaporation pattern on the grid suggested that it was UHF
>> current - not VHF.
>
> Sure, high UHF/SHF oscillation *NOT* the VHF parasites you claim
> to cure with resistance wire.

In 3-500Z amps, the parasite is typically below 150MHz.  In 8877 amps, 
the parasite is apparently in the UHF region - judging by the 
evaporation pattern on the grid.
>
> 73,
>
>    ... Joe, K4IK
>
>
>
>



Rich Measures, 805.386.3734, AG6K, www.somis.org

_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps

<Prev in Thread] Current Thread [Next in Thread>