[Amps] Follow-up...re: Time to ask for help.

Mon Jul 27 16:49:38 EDT 2015

"Grounded screen" will work as it did in the Collins 204F-1 and the 208U-3/10 amplifiers,which were neutralized
despite using liberal grid swamping.

The electron transit time input losses at 50 MHz for the 4(C)X150/250 can be approximated with a series resistance of about 5 ohms in series with the grid input capacitance of 20 pF.

This translates roughly to a parallel equivalent of 4000 ohms and 19 pF.
4 tubes will correspond to 1000 ohms and 76 pF,and the grid voltage for AB1 will be about 55 V peak.

With reasonable coupling and circuit losses, the required peak driving power is around 2 W.

Output power should be about 1000W,so the power gain is 37 dB, which is way above the 20 dB
that are considered "safe" even with neutralization, and assuming no other feedback paths.

A more reasonable safe gain without neutralization would be in the order of 15 dB, which requires 
about 100 ohms extra swamping.

73/
Karl-Arne
SM0AOM

----Ursprungligt meddelande----
Från : mike at tubby.org
Datum : 2015-07-27 - 21:01 (V)
Till : amps at contesting.com
Ämne : Re: [Amps] Follow-up...re: Time to ask for help.

I've looked at the schematic.

As you have SK640 sockets why not modify the design to run "grounded 
screen" ... like the 144MHz 4CX1000A design in the 1990 ARRL handbook?

You would connect B- to the cathode, and the +ve side of the screen 
supply to chassis and the -ve side of the screen supply to cathode. You 
would need some good quality (large) silver mica caps to return the 
cathode to deck at RF.

In this mode you have a 'perfect' RF ground on the screen and the 
isolation between grid and anonde circuits should be at least 10dB more 
than the stage gain and as a result you shouldn't need any neutralizing 
at all and you can maintain your pi-tuned input and pi-tuned output.

To execute this approach successfully you would probably need to use 
some fingetstock to fabricate a low inductance ground for the screen 
contact with multiple earth points.

Just a thought ...

Mike G8TIC

On 27/07/2015 19:45, sm0aom at telia.com wrote:
> If you are using the SK-640 sockets, which lack integral screen bypass capacitors,
> what kind of other screen bypass do you use?
>
> Be advised that attempting linear operation of the 4X150 on 50 MHz using conventional screen bypassing
> is a disaster waiting to happen. There is very little control over the self-resonant frequency of the screen circuit under these circumstances, and a "cold screen grid" is essential for stable operation.
>
> The previous suggestion of operating the amplifier in "grounded grid" will only work if the proper voltages and bypassing are present on both grids.
>
> Just tying both grids together and to ground will easily damage the control grid, as the 4X150 family is not suitable for operation as an "high-mu" triode.
>
> 73/
> Karl-Arne
> SM0AOM
> SM0AOM
>
>
>
>
>
> ----Ursprungligt meddelande----
> Från : ai7rogerroger at gmail.com
> Datum : 2015-07-27 - 20:17 (V)
> Till : donroden at hiwaay.net, amps at contesting.com
> Ämne : Re: [Amps] Follow-up...re: Time to ask for help.
>
> Sockets in use are Eimac SK-640.
>
>
> 73, Roger
> AI7RR
>
> On Mon, Jul 27, 2015 at 10:49 AM, <donroden at hiwaay.net> wrote:
>
>> I woke up and the thing on my mind was .....
>>
>> Are you using Eimac SK-630 sockets ?
>>
>> I think those sockets may have been the reason my amp was so stable on six.
>>
>>
>> Don R.
>>
>>
>>
>> Quoting Roger <ai7rogerroger at gmail.com>:
>>
>>   Don,
>>> Thank you for the information. I've been studying other circuits and
>>> combined with the info you just mentioned regarding screen loading, I have
>>> a change to make. Both my grid and screen supplies are very stiff with
>>> capability of supporting reverse current. My screens however are shunted
>>> with very high resistance at the terminals just to hold them at ground
>>> during rx.
>>>
>>> I've gone through many different wiring configurations with this amp and
>>> have even had output once or twice, but every time I try to improve it, I
>>> screw something up. This time finding the error hasn't been so easy.
>>>
>>>
>>> 73, Roger
>>> AI7RR
>>>
>>> On Sun, Jul 26, 2015 at 8:17 PM, <donroden at hiwaay.net> wrote:
>>>
>>>   No problem... Just think of the 250Bs as a big variable DC load on the
>>>> plate power supply.  You should be able to change the grid voltage and
>>>> see
>>>> a smooth
>>>> increase in plate current from zero to near 1000ma ( very briefly !!! )
>>>>
>>>> I would suggest zero to 500ma as a minimum range to check.
>>>> No more than that until absolutely sure everything is stable
>>>>
>>>>
>>>> THEN hook up the grid circuit and see if you can do the same thing..
>>>> something may be breaking down in the grid circuit and killing the bias
>>>> voltage and
>>>> then ...... BOOM ... Plate current runs away.
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> Quoting Roger <ai7rogerroger at gmail.com>:
>>>>
>>>>   "I'm confident my DC is stable."
>>>>
>>>>> Scratch that. I misunderstood the statement:
>>>>>
>>>>> "He has to make the amp DC stable before hooking up the input and output
>>>>> RF
>>>>> networks."
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> 73, Roger
>>>>> AI7RR
>>>>>
>>>>> On Sun, Jul 26, 2015 at 5:50 PM, Roger <ai7rogerroger at gmail.com> wrote:
>>>>>
>>>>>   Sorry, I thought it was clear what I was trying to do. I have four
>>>>> 4x150a
>>>>>
>>>>>> tetrodes in parallel, attempting to make them operate on 50mhz. I'm
>>>>>> confident my DC is stable. Additional information includes, a Pi tuned
>>>>>> input and output. Input and output terminations are quality 50 ohm
>>>>>> devices.
>>>>>> Power supply and rf deck are in separate boxes. My problem appears to
>>>>>> be
>>>>>> related to instability possibly caused by insufficient or incorrect
>>>>>> neutralization.
>>>>>>
>>>>>> It was mentioned that the neutralizing cap needed to be by-passed and
>>>>>> that's something I have not even considered. By-passing anything in the
>>>>>> grid circuit seems contradictory to me.
>>>>>>
>>>>>>
>>>>>> 73, Roger
>>>>>> AI7RR
>>>>>>
>>>>>> On Sun, Jul 26, 2015 at 3:56 PM, <donroden at hiwaay.net> wrote:
>>>>>>
>>>>>>   Quoting Ron Youvan <ka4inm at gmail.com>:
>>>>>>
>>>>>>>
>>>>>>>    I am trying to find out what you are trying to do.
>>>>>>>
>>>>>>>>
>>>>>>>>>    Me Too Ron,
>>>>>>> He has to make the amp DC stable before hooking up the input and
>>>>>>> output
>>>>>>> RF networks.  We don't know if that's the case yet.
>>>>>>>
>>>>>>> Don W4DNR
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Amps mailing list
>>>>>>> Amps at contesting.com
>>>>>>> http://lists.contesting.com/mailman/listinfo/amps
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>
>>
> _______________________________________________
> Amps mailing list
> Amps at contesting.com
> http://lists.contesting.com/mailman/listinfo/amps
>
> _______________________________________________
> Amps mailing list
> Amps at contesting.com
> http://lists.contesting.com/mailman/listinfo/amps

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