[Heath] AR-3 Tuning Cap

[Kenneth G. Gordon]

On 19 Apr 2016 at 17:39, Al & Dottie Skierkiewicz wrote:

> The capacitor does not contain "gear reduction" in the true sense of the word.

Correct.

> The inner shaft connects to the outer shaft through ball bearings.  As
> the inner shaft turns, the balls turn and then so does the outer
> shaft.  The reduction is the difference in the diameter between the OD
> of the inner shaft and the ID of the outer shaft.  

Yup. And the balls are what constitutes the "planetary transmission". Works on the exact 
same principles as the Model "T" Ford transmission....and most automatic transmissions 
today.

> If the lubricant has hardened, the inner shaft and outer shaft spin at
> the same time, no reduction.  Most of the reduction drives operate
> this way.   

Yes, although some of the very early vernier dials worked on a bit different principle: a small 
wheel operating against the rim or in a slot on a larger wheel, the ratio being basically the 
difference in diameters.

At this point, I think I am going to dismount the tuning capacitors I have here, turn the rotors 
so that they are vertical so that I don't damage the bakelite separator on the rotor, then 
immerse them in PB Blaster until the shaft is covered, then let them sit until enough lube 
leaks inside to "reconstitute" the original lube.

If anyone has a better idea or one which might work better, please let me know.

> I am rebuilding an AR-3 right now.  Much of the receiver uses the tube
> sockets riveted to the chassis for ground. 

Riveted? That's kinda weird. All kit-built AR-3s used nuts and bolts to hold the sockets in 
place.

> As the receiver aged, these became high impedance and therefore ground
> bypass circuits became less effective. 

Which is a VERY common problem with all old receivers with riveted sockets and terminal 
strips.

> That includes the tabs on the power supply filter.  I also experience
> the hum in BFO position that varies with RF gain.  

Yes. Mine still does that, but at least now I know what to look for: bad grounds, mainly.

My very first receiver, an AR-3, didn't have a working BFO at all. That was very difficult to 
use to copy CW.

> I once owned an AR-3 when I was in high school.  I purchased it used from the Radio TV Lab on 
> Irving Park in Chicago if you lived on Chicago's north side.

My Wife is from Illinois and worked in Chicago: I'm from Montana.

> I found one at a Milwaukee hamfest a few years ago, I didn't even
> haggle the $30 price tag.  When I got it home it made hum but I could
> only pick up WGN radio, the closest BC transmitter to my house.  After
> a little troubleshooting, I came to the conclusion the receiver had
> been built but never aligned.  I found some wiring errors as the
> builder kept doing mirror image translation when wiring from the top
> of the chassis to the bottom.  With alignment, things started picking
> up but as I worked, one day I would have sensitivity and the next day
> almost nothing.  It was then I discovered that whacking the chassis
> and then the IF cans in particular, changed the sensitivity.  (this is
> referred to as "thunder" in some rebuilding publications)  With a
> little research, I found the dreaded silver on mica caps in the IF
> cans. 

Oh, crap! Well, my IF cans SEEM to be OK so far: at least they align easily... I'd better 
check them though. Thanks for the warning.

> Yes this one had the silver that migrated to other parts, high
> impedance connections in the clamp, etc.  I am rebuilding the IF cans
> with silver mica dipped caps and hope for better performance.  This
> part of the project is taking longer than expected as I also mentor
> high school robotics and that is taking most of my free time. 

Most of us understand such matters. :-)

> I recently purchased a QF-1 Q multiplier for this radio.  The "mirroring " problem also hit on the 
> accessory socket so I had some rewiring there as well.  I cannot get the Multiplier to function at 
> this point in any function.  I do get noise and hum when I switch but that is it.  Strange as it may 
> seem, the same person may have built both units.  I replaced the RG58 to the IF stage and I get 
> all the right voltages.  I have never used a Q device so I am still learning on that.  My suspicion 
> is that the IF is so far off frequency due to the bad IF caps, that I may be out of the passband 
> with the Q Mult.

That sounds plausible. I have several QF-1s here, and they all work very well indeed. I use 
them with many different receivers, changing the tuning enough to center it in the IF 
passband when necessary. I use a somewhat modified QF-1 with a Scott SLR-F, for 
instance, which has a 465 KHz IF. After reviewing many different "Q-Multiplier" circuits, in 
my opinion, the Heathkit QF-1 is one of the better implementations of the idea.

BTW, I am not sure why Heathkit used a short direct connection to the plate of the first IF 
stage so that the connection to the QF-1 had full plate voltage on it, but I intend to find out. 
Seems to me that a series-DC-blocking cap would make life easier for those of us who use 
the "Q-Multiplier".

Also, don't forget to try the 12CS6 in place of the 12BE6. Doing so here made a significant 
improvement in sensitivity. My AR-3 still has a long way to go to achieve peak performance, 
though.

Ken W7EKB