In <199611181952.LAA29420@yvr.cyberstore.ca>, on 11/18/96
at 11:52 AM, eldridge@whistler.net (Bob Eldridge) said:
>Some questions about MLA2500 operation on 160:
>3. But why is the Q said to be low anyway? C9 is 300 pF across the 140 pF
>variable, and <440 pF should give a Q of at least 10 shouldn't it? I do
>agree that to resonate properly on 160 this C has to be increased,
>because the efficiency goes down if turns are added to the coil.
The plate impedence of a tube circuit is variously V/(2xIp) or
V/(1.57xIp) depending on class of operation (use 2 for Class C, 1.57 for
AB) where V is the plate voltage and Ip is the peak plate current. For
the MLA2500 this works out to be about 1600 Ohms (based on 2 KV and 800
MA, which represents 600 watts of plate dissipation at 60 - 65%
efficiency) at 1 KW (peak) out.
The maximum plate tuning capacitance of 440 pF represents Xc=200 Ohms. Q
is Z(plate)/Xc which makes the MAXIMIM Q of the plate circuit 8 (at 1.8
MHz). To achieve a Q of 10 requires a tuning capacitance of 495 to 560 pF
(add another 120 pF across C9).
Using the same constants (Q=10 and Zp=1600) yields a required maximum
loading capacitance of 2600 pF at 1.8 MHz. It appears that the stock
loading capacitor (45-500 pF in parallel with 2100 pF) is marginal at 1.8
MHz even with a perfect 50 Ohm load. One might consider adding another
200 pF across C47 ... but that might be too much loading cap at the top of
the band.
The MLA-2500 requires a flat load to tune properly ... there is not enough
range in either the tuning or loading capacitor to handle a mismatch.
Does anyone know what the combined value of L1+L2 (the plate tank) are in
circuit, particularly with the presence of the top and bottom covers?
>5. Has anyone tried relocating the tank coil downwards, away from the top
>cover? Or modifying the top cover? Raising the cover increases power
>output (although my top cover doesn't get warm running what I consider to
>be normal output - 600 watts).
Moving the tank coil downward will move it closer to the bottom cover and
probably not gain anything.
>Excessive grid current kills 8875s, and I find the optimum condition is
>when drive and loading are adjusted to just barely move the grid drive
>meter. The extra output resulting from more drive just isn't worth the
>risk.
The MLA2500 needs grid protection ... one suggestion has been to limit the
grid current to 80 mA for a pair of 8875s and that seems about right. I
don't know what this works out to in terms of matched power output but we
should be relatively comfortable in running the tube at 1600 W input (1000
W peak output) if the grids are properly protected.
--
73,
... Joe Subich, AD8I
<AD8I@ibm.net>
<subich@ibm.net>
--
FAQ on WWW: http://www.contesting.com/topband.html
Submissions: topband@contesting.com
Administrative requests: topband-REQUEST@contesting.com
Sponsored by Akorn Access, Inc & KM9P
|