[Amps] 3-500's and stability
Thu, 22 Aug 2002 21:16:20 -0700
> I'm sure I've posted this before but for the benefit of the new
>guys I'll do it again.
> I have on old Drake L-7 that has been my workhorse for many, many
>years. The amp is extremely stable. There has not been one pop, snap
>crackle or bang as far back as I can remember. I've used it from 160m
>thru 10m including WARC bands (30m into dummy for test) without a
>problem. What is the secret? I installed a QSK system based on ideas
>from Rich's website. Why is this good? I have a decent dual trace
>scope which I hooked up to monitor RF input and output for testing
>before and after the mod. The original old 3 pole TR relay is WAY too
>slow! The TS-940 exciter starts producing RF about 6ms after keydown.
>The old relay hasn't even started to budge at this time! The scope
>shows the 100 watt output of the 940. At about 12ms the relay has
>closed but the contacts are bouncing. The scope shows that the RF
>is cutoff during the contact closing interval and then there is a
>God-awful mess as the contacts bounce, connecting and disconnecting
>the input and output. How any amp could survive this without
>triggering some sort of nasty oscillation is beyond me. Pure luck I
>think. With the QSK system, all is sequenced nicely. No hot switching
>or nasties anywhere.
** Thanks for the t measurement on the 940. 6mS is quick. I agree
that hotswitching produces the kind of anode-current transients that are
ideal for ringing the anodes' VHF-resonant circuit between the tube and
> Two other aspects. I directly grounded the grids with copper strap. No
>more resistors or RF chokes. Is this good for stability?
** For best stability, the grid resonance should be as high as possible.
I found that there was only a couple of % difference in grid-resonance
between directly grounded grid and C grounded grid -- with the C gnd
being a bit higher in freq.
>I don't know
>for certain. I know it never solved the problem in any of the other
>amps I've worked on. The QSK system allows you to separate the input
>and output relays. I'm not certain if this helps either but I've always
>been told to keep inputs and outputs separated to avoid oscillations.
** There is virtually no VHF at a DPDT antenna relay even when the
amplifier is inadvertently making VHF. The reason: the HF tank is a LPF.
> Mine is the only Drake I've done but I nearly went crazy trying to
>tame friends TL-922's. The only thing that worked was the QSK system.
** QSK helps but the beautiful, high VHF-Q silver-strap "parasitic
suppressors" are a contributory factor.
>I'm sure that there are other relay sequencing schemes out there but
>QSK sure is nice for CW ops!
** I use it for SSB-VOX because it makes about 25db less accoustic noise.
>Amps mailing list
- R. L. Measures, a.k.a. Rich..., 805.386.3734,AG6K,