[Central-Electronics] 100V restoration notes

[GRANT YOUNGMAN]

I’ve posted a few times lately about the wide range of continuing issues I had getting my once-working 100V working again.  I’m happy to report that after about 4 weeks of on-again-off-again effort it is finally working well, back in the case, and integrated with an SX-115.

If you end up in a similar boat, all I can say is that a doggedly rigorous approach is the only way to guarantee results.  The problems ended up being a combination of several things — (1) a couple of tubes (1st mixer, driver) that even though tested good did not work well in the radio, oxidation (on coil box pins/connectors) that did not respond to simple cleaning and application of D5L, and alignment errors that can creep in even if you use the specified interfaces in the manual.

At one point I installed NOS Sylvania 12BY7’s in the first/second mixer.  The new tube worked fine in the 1st mixer, but not in the 2nd.  I ended up with the original Sylvania in the 2nd mixer position — which was not a small thing, since the “used” tube nearly doubled the output of the NOS tube.  The manual suggests that if you can’t test each individual tube, you should replace all of the tubes in the RF chain.  But even new tubes that individually tested well didn’t produce the best result, and shotgun replacement didn’t either.  The only way to test a tube in the RF chain is to put it in the radio and measure RF output.

Many of the bandpass RF circuits are VERY sensitive to the presence of a probe in the various test points, which I can only surmise upsets the capacitance on the secondary side of the tuned circuits — even the recommended probe circuits in the manual did this.  As a result, it was necessary to tweak the tuning after the fact using RF power out as the only reliable measure.  This will disrupt the pretty flat bandpass curves you get using a sweeper as specified in the manual, but I finally concluded it was better to tweak as necessary for maximum in the amateur bands, and not worry too much about the frequencies outside of the bands.  Also, and it isn’t mentioned in the manual, don’t forget to tune the 8Mhz inductors for max RF power out.  It occurred to me that perhaps a final sweep end-to-end alignment using 5.5-6.5 Mhz injection at the VFO input and the scope on RF output might be an idea to explore, but I haven’t gone there at this point.

I finally had to do two things with the coil boxes.  First, I gently tightened all of the socket connectors (the ones that are reachable), and second, I ended up using 1000 grit paper from my plastic polishing kit (lightly) on all of the pins on the cans and one gentle swipe of a small tightly rolled tube of the same 1000 grit paper into each connector contact.  I used zero residue contact cleaner to remove all trace of D5L from the sockets and pins.  This finally solved the issues I was having with intermittent behavior of some tuned circuits in the inductor cans. 

There were two things left undone.  The bias on the 6BQ5 driver tube is low, but the resistor divider that sets it is buried under the bandswitch. The 200V has a pot, but the 100V does not.    I never found a way to get to it without risk of doing damage around the part.  Also the adjustment of the pots to minimize spurious output from the two mixer stages did almost nothing to the spurious signals (which were already substantially down), but had large effects on power out.  The pot in the cathode of the 2nd mixer had no effect on anything.  Maybe another problem to be worked at some point in the future.

And I have to thank Bob Sullivan, who argued that taking the radio to a gun range and using it for target practice was perhaps not an effective restoration method :)

Grant NQ5T