[R-390] Gain drift

[dog]

I think I'm getting somewhere now. Put the Boonton on pin 5 of V504 4th 
IF amp. Input 54uV into the IF module because that was all it took to 
get ~600mV out of V540 which is near the top range on the Boonton. 
Checked with a 10dB smaller input and same 10dB on V504 plate. So it's 
not compressing. After a few minutes the rf volts start dropping off, 
I'm going to step back to the grid of 504 and see what's going on there 
next. AGC is disconnected. I was getting a few volts on tb102-3 and it 
didn't change when the plate of V504 dropped. Not sure what that means. 
But I see the AGC is coming from the grid of V504 to a cathode follower 
and the AGC IF amp comes off that. AGC rectifier comes off the plate of 
AGC amp. Looks like the 4th IF amp V504 is not controlled by AGC.

C547 and C548 are replaced.

Reason I didn't change C551 is I read that it wasn't usually a problem, 
but looks like maybe could be. I've got a 2MFD NTE film I think, it 
looks like a large orange drop but it's burgundy colored. I'll have to 
look back in my Mouser order to see what it is. That will fit in that 
C551 can, but I'm going on with the gain in V504 for now.

The 2.7M is R544. I haven't replaced it. Yet. It's in the plate circuit 
of the AGC time constant.


On 10/19/2018 20:16, Larry H wrote:
> Dave, The idea for disconnecting the 3/4 link is to allow the agc 
> circuitry to work but not affect the IF gain. This will provide for a 
> constant gain (if all is working correctly) for easier testing.  
> Although grounding # 4 creates max gain in the IF, if less gain is 
> desired, the RF gain controls the gain of the 1st 2 IF tubes.  The 3rd 
> IF gain is controlled by the internal IF gain adjust.  If C551, C547 
> and C548 are original, I'd look very closely at them.  C551 can leak 
> like a normal cap, but it can also leak to it's case, which is gnd.
>
> While testing, I suggest using the lowest level of input as possible, 
> as it is possible to overdrive the 4th IF and obscure your results,.  
> Although the IF should handle input creating a diode load of 15 vdc, 
> 7vdc is safer.  The issue is driving the 4th IF or the agc amp into 
> clipping.  This can happen in mgc mode, but in agc mode the agc helps 
> prevent this from happening.  Since the agc is disabled for testing, 
> you need to be careful of this.
>
> Regards, Larry