[R-390] an audio deck saga (long)

[wli]

Recently I had the occasion to do a ground-up audio deck restoration. This unit appeared to be a St Julians survivor with some surface rust and dirt. I used the excellent schematics in the Y2K v3.0 manual as my reference, the many posts from members of this list, as well as recent articles in ER re audio deck modifications. 

Inspired by Nolan Lee's description of his compulsively complete EAC restoration a few years back, this time around, I decided to approach the project as if I had all the time in the world. So I totally stripped the unit down the chassis and threw everything into a bucket. First checked every component, wire, and socket during reassembly.  Rust and chipped paint was dealt with first (easy part). 

I found a few loose ground lugs, lots of deteriorated solder connections hidden under the black insulation on J619 and J620, and many loom wires coated so bad under the insulation as to not take solder.  Near all of the carbon resistors had drifted up as expected, but none of the 50 year old Vitamin Q capacitors from General Instruments showed any appreciable leakage at 250v. Luckily all the chokes and transformers had their DC resistances very close to specs, and none had any leakage to the case. All the tube sockets were discolored (overheating?)... so new ceramic sockets went in their place. New electrolytics were stuffed into octal relay cases. Most all of the small components were replaced with new or checked units. I ran a single copper bus down the middle and grounded it securely at one point. Known good tubes from my stock were installed.

Having rebuilt it as well as I could, I was happy to see it closely matched resistance and voltage parameters seen in my other units; and, when powered up.... worked. So far, so good. 

Then, I got to thinking about improving the audio. Since this receiver was basically designed as an intercept unit, I wondered what could be done improve its performance in just that area.

Ray Osterwald wrote a nice history behind the audio deck back in 2004 (ER vol 181 pp45-46). Seems a lead Collins engineer named HE Houge spent some 3000 hours designing the deck in 1949 to meet the requirements set down by the US Signal Corps. A large amount of negative feeback thru R612 was employed not for the usual reasons of dropping harmonic distortion and improving frequency response etc... but to meet the Corps specifications re output impedance. The resulting gain loss was treated by adding positive feedback through R615.

Chuck Felton published his audio modifications (ER vol 183 pp 7-10) which made interesting  reading. 

There has been an enormous amount of comments and experience from this group dating back to 1997 re audio mods, which I looked over again. Armed thusly, I went ahead and tried a few simple easily reversible changes to the now functional deck. The rationale behind these moves have been outlined by others. 

a) removed R612
b) dropped R615 to 24 ohms (probably should be removed) 
c) removed C609 (the troublesome small wet tantalum electrolytic)
d) added a 10M resistor from  plate of V601A to grid of V601B
e) kept the original 6AK6 and T601
f) shorted R101 (to get rid of the voltage divider) 
g) ran the local output to a surplus 600 to 8 ohm transformer mounted in a Navy aluminum speaker box (pseudo LS-206...); a half cubic ft aluminum box with a 6 inch car speaker. Looks swell, with ridiculous acoustical properties.

End result was short of amazing for such small changes. Audio was louder, with intelligibility clearer probably due to limited frequency response. No squeals or hum at any gain setting. I suspect that fixing the various grounds, replacing all the out-of-spec resistors, replacing various wires, resoldering everything, and using known good tubes contributed a lot to the final result. Anyway, for speech, I found best results using the 8KC filter.

I confess I originally had the intention to perform the Kleronomos audio addition, but upon mature consideration, it was  easier to just run my old single channel 20 watt Williamson (6L6GB's) amp off of the diode load jack, whenever I really wanted great hi-fi. 

Obviously, I could have gotten *radical* by employing a 6AQ5 or 6BQ5 in place of the low power 6AK6, or installing a nice Hammond 600-8ohm transformer etc etc; but what I wanted to see, was if any improvement could be achieved just employing small parts. 

I think it can.

Thanks for the bandwidth.. duckin' and running....

W. Li
Mercer Island