[R-390] 6AQ5 Mod??

[Roy Morgan]

At 06:40 PM 9/2/02 +0000, you wrote:
>Hello All.
>Working on the 2nd R-390A (Dons machine).  It has one dead and one very
>very weak 6AK6 in the audio section (explains why the Line audio didn't
>work.
>
>I remember last time I rebuilt a R-390A I subbed 6AQ5's for the audio
>output stages.  If I recall correctly all it entailed was changing the
>wiring on a couple tube pins (which I can figure out from the tube
>manual).  Checking around I no longer seem to see this mod on the web.


Here are my notes from WB2ADT:
I also have his schematics and other stuff..
Got this from some web page.



Introduction

The audio section in most receivers seems to be at best an after thought in 
the overall aspect of radio design. The R-390a is certainly not the worst 
of the bunch but
can stand an improvement. Its definitely better than today's $2000.00 rigs 
with there three inch squeakers. Also having dual audio amps is one aspect 
of these
receivers I've always liked. The line level section with its VU meter is 
perfect as is for driving sound cards, tape recorders, and such. This 
section was left as is.

So why bother with widening the bandwidth and reducing distortion? AKA 
improving fidelity. Some have the opinion that it should be left as is 
since the R-390a is
after all a communications receiver and such things should have that touch 
to it.

My opinion is otherwise. We're so used to listening to crummy sounding 
output that it's become some type of a de-facto standard.

In receiver design its been pretty much a case of the guys designing the 
bullet proof frontends, highly selective IF stages, and signal processing 
circuits that get all
the glory. The audio design is relegated to a novice engineer whose been 
instructed to use an existing, off the shelve module or IC and see what he 
can do with it.

So what can we expect from your new modified audio stage?
- Listening pleasure, high distortion and narrow bandwidths produce 
listening fatique. Reducing it will keep you listening longer and enjoying 
it more.
- Honest signal reports, ever notice that the "sound" of all the signals 
seem to be about the same? Opening up the audio can produce some 
interesting results; e.g.
Biff in Northern NJ has a pretty lousy sounding signal. Freddy down in 
Jonas, PA ssb signal seems about the same. Well, after the modification 
Biff's signal is
worse than thought and Freddy should get an award for having a really 
quality sounding signal.
- Same goes for commercial and shortwave stations. Some could use some work 
and others have a clean signal. There's some really high quality programs
produced on shortwave. Take advantage of it.

So on with the modification! Firstly a rundown of my self-inflicted rules.

- Any changes should be 100% reversable.
- Any new componients must be affordable and easily obtainable.
- No new spares to be stocked.
The last I wasn't too lucky with. You're going to have to put an extra tube 
on the parts shelve.

Following is a rundown, stage by stage, of the changes. All the changes 
have some sort of reasoning behind them. Some look like maddness and I 
really at times
can't explain what I was thinking, verything does work well though. Here 
they are.

1st AF Amp (V601a)

A 5814/12AU7A wouldn't be my choice as an audio tube. Right out of the box 
its distortion is high. Things can be done to get it to an acceptable 
level, say 1-2%
The existing feedback loop was removed, this would be C601, R602. C612 can 
be removed if you have it in your audio deck. This was a mod in later 
receivers. If
there its to boost the treble. Interesting, seems someone was trying to get 
a little extra out of this amp. C609 is taken out. It's just a hindrance 
for better bass, as is
C602 which is replaced with a .1 uf 250 volt polystyrene capacitor.
The bias on this tube depends on your actual voltages produced by the power 
supply in the receiver. Some sets have been solid-stated, some aren't and 
powerline
voltages are all over the place. That's where the beauty of self-biasing 
comes in. But we're looking for less distortion. If one runs a 12AU7 at 
about -4 volts on the
grid its a pretty clean sounding tube. So to hold it there I opted for 
fixed bias. How to get it? Easy. Put two LED's in series and use them to 
replace R604. Result is
there's always -3.9 to 4 volts on the grid. The LED idea was someone else's 
bad dream, not mine. Works great!

AF cathode follower (V601b)

Didn't find anything to improve. No changes here. Just gives the other half 
of V601 something to do. I would have left this whole stage out, then again 
I didn't do
the original design.

Local AF Amplifier (V602b)

I found the design of this stage to be rather interesting. A very high 
value plate resistor is used, along with a rather large amount of negitive 
feedback. Also a small
amount of regenerative feedback is employed also. I had to ponder about 
this setup for awhile before I remembered that some amplifiers used 
positive feedback to
cancel out distortion by working one tube curve against the other. Not a 
bad idea, except in practice, between parts variation and aging, it never 
quit worked out to
well. This technique was rediscovered several years ago by a tube amplifier 
designer, but has been around for 50-60 years, just not used to much. 
Anyhow R611
was replaced with a 56K resistor, R612 was removed, so was R615, and a 
jumper put in its place. Two LED's in series was used again for fixed bias 
on the
cathode, eliminating R610. A 1.0 uf capacitor was substituted C605, which 
completes this stage.

Local AF Output (V603)

Here's where we ran into a sticky wicket. Replacing the transformer was one 
of the first changes I did. It was a better quality unit that matched my 
favorite 8 ohm
speaker. It had a wider bandwidth. Boy, did the old 390a sound bad. This 
was before any other changes had been thought up. Opening up the fidelity 
brought up
the old audio amp adage, high distortion, narrow the bandwidth. It works. 
In the original setup. After going over the output stage and even 
breadboarding it I
couldn't get the distortion lower than 12%. Finally, after putting out a 
call on the 390a list I got my hands on the tube curves and some addition 
specs on the 6ak6.
16% distortion is the norm for this tube, about 10% with some feedback. As 
much as I tried I just couldn't get it down to where the audio was 
listenable. I even
tried a single ended setup driving a push-pull transformer. This requires 
operating on the transformer using one primary winding on the plate and the 
other primary
winding on the cathode. Interesting way of setting up an amp but it didn't 
do a thing for it. I also went against one of my rules of keeping it 
simple. Splitting a
primary winding is no piece of cake. So I went on search for a better 
output tube.
After much diliburation I choose the 6aq5. Wiring it as a triode keeps the 
distortion down, about 2 watts output is plenty for most listening, no 
opening up the
chassis for a nine-pin socket. It's still plentyful and cheap. Need an 
extra spare though. Oh well, nothing's perfect.
Ok, so basically we rewire the socket to accomodate the 6aq5.
Remove the screen voltage wire from pin 6, insulate, and tuck into a safe 
spot in the harness.
Clip the wire going from pin 2 to pin 7.
Move the remaining wire on pin 7 to pin 2.
Move the wire from pin 1 to pin 7.
Add a 100 ohm 1/2 watt resistor from pin 5 to pin 6.
Remove the 6ak6 if you haven't done so already and replace with a 6aq5.
Remove R614 on terminal board and put a 15 volt zener diode in its place. 
That's about it for the output stage.

Transformer

My final selection was one out of my Junque box/room. You might have 
something around, if not the tranformer in the parts list should be OK.
Mounting was done by breaking off the left-hand tab looking at the primary. 
It was then bolted in diagonally and then wired up into the harness.

Speakers

I've tryed several so far. If it sounds good on your Hi-Fi it will do OK 
with this audio change. Although 2 watts doesn't sound like alot I've 
driven a small
bookshelve speaker system that is a 2-way using a 6 1/2 inch woofer to 
plenty of volume. I also have a 12 inch full range speaker mounted on a 2 x 
4 foot baffle.
My favorite one is a Radio Shack 4 inch full range that's mounted in a 
5x7x4 inch sealed box. This is one good sounding setup.

- Roy Morgan, K1LKY since 1959 - Keep 'em Glowing!
7130 Panorama Drive, Derwood MD 20855
Home: 301-330-8828 Work: Voice: 301-975-3254,  Fax: 301-948-6213
roy.morgan@nist.gov --