You might want to check a couple of things.   For starters, if the rig keys, but you have no power indicated and if your rig otherwise works with a microphone, then you probably have too little audio from the USB interaface into the rig.

Make sure that your Windows system hasn't selected the USB interface as the default sound card, to prevent every Windows beep and noise from being transmitted.  Use the computer's built in sound card interface as your default card for Windows sounds and setup only your ham radio software to use the USB interface.  

Then go into your sound card settings (Control Panel/Sound) and set the software Playback and Record volume for the USB interface to a level that works for your rig.  Sometimes it works better to just set the software controlled volume settings to 100% and use the knobs on the interface box to set the level needed by your radio.   Start with the software Playback set at 100% and the TX knob at 50% and try adjusting it from there.  Once you get it working, do not overdrive the rig.   Also, turn off the speech processor on your rig.  

Let us know the symptoms after you try that and we'll see what else we can suggest.

Bob

I wouldn't worry about the phrase "laying around" when they talk about wire.  An MSS will probably have a lot of old wiring harnesses in boxes.  They were sufficient for 100W Motracs, so the wiring will work fine for your purpose.  As the other poster indicated, any MSS worth its salt will install your radio correctly.

In my experience with the 6BTV (and I've had several), you probably have enough radials down.  Most people find that 32 is about right, with only marginal improvement when adding more.  Here in the extremely dry desert soil of Phoenix, I used 60 radials at 55 feet each, but that's probably overkill for most installations.  In my first installation years ago I thought I could get away with no radials since the antenna tuned very nicely, but my station was as deaf as can be.  My current installation is easily the best I've ever had and I could not be happier.

As you add radials you will probably find that the SWR curve gets narrower and you will have to shorten the antenna.  Those are signs that you are doing things right.  You may have to shorten the traps about 1/8 to 1/4 inch to achieve resonance in the center of each band.  You should not cut any of the tubing.   For those who might want to read up on this antenna, the DX Engineering web site has several articles available, including info on exactly how to shorten the traps, install radials, and tune the antenna properly.  Their info is a bit better than the instruction sheet provided with the antenna and I learned quite a bit from them.

Can you elaborate on the symnptoms?  The actual sound card sample rate doesn't drift.

The rig frequency might drift a bit.  I'd try to isolate the problem a bit more before spending money on another sound card.  I've found the SignaLink to be a very good interface and I seriously doubt that another sound card would work any better.

The writers probably thought the words sounded high-tech and futuristic.  Clarke's third law is that any advanced technology is indistinguishable from magic.  I kind of feel that way about quarks, leptons, and Higgs bosons.

The Enterprise didn't return to Earth because they couldn't find dilithium here.  They did search the universe for dilithium crystals though, at least until Spock figured out how to synthesize dilithium in "The Voyage Home".  The search provided a reason to beam down and get into fistfights with aliens.  

I always wanted to have a ham shack that looked like the bridge of the original Enterprise.  So many flashing lights....  ;-)

I agree with K6AER.  Eimac says filament voltage has a big impact on the life expectancy of the 3-500z.  A few tenths of a volt can make a difference in tube life.  The filament voltage is highest when the amp is unkeyed and drawing the least amount of line current.  A drop in filament voltage under load will not have anywhere near the effect on output power as the drop in HV does and the filament current doesn't increase when the amp is keyed.

Phil also offers good advise (as usual) when he advises  to check on the correct side of the filament choke, assuming you still want to do this.  It is there to block the RF.  

Are you trying to solve a specific problem with the amplifier?  Perhaps we could help with that.

A padded wooden block or a box set next to the piece you are working on can provide a rest for the pinkie side of your palm or wrist while soldering.  A family member finds that when he applies a bit of resistance to his hand with a large rubber band that it lessens the tremors a bit.  He uses a piece of velcro around his hand and attaches a rubber band to it and to something heavy.  I'm not sure of the exact details, but for him it doesn't take much resistance.  He doesn't know why or how it works.

I wish you the best.  

73,
Bob

This link also discusses the deviation adjustment from 5 KHz to 2.5 KHz.

http://www.radiomods.co.nz/icomic240mods.html

To answer your other question, the SB-220 used eight 200uF/450v caps in series, for a total bank capacitance of 25uF.  You can use larger caps, but as other have noted the initial inrush current to charge a larger capacitor may be a bit high, so a step-start circuit is advised.

<The rumor, mythology I'm going on is to replace those ole caps before, well, before ?>

It's not really a myth.  Old filter caps DO dry out over time.  Sometimes they fail in a nice way and sometimes they scare the heck out of you.  I've had a metal can capacitor zing past my ear and punch through the ceiling tiles, so I'm a believer.  The problem is that they seldom give any warning before they fail so it's hard to know when they have gotten "too old".

IMHO, the biggest problem with the SB-220/221 capacitors was that builders would not properly space the bleeder resistors away from the ends of the caps.  That left no room for convection to cool them properly.  The heat buildup would melt the plastic holders and would also damage the capacitors over time.  As a practical matter, if the caps are not leaking (and you can usually tell by looking at them) they seem to last a long, long time.  

The cautious and nervous side of me always wants to replace electrolytics.  My practical side tells me that there are many thousands of these amplifiers being used by capable hams every single day with no trouble at all.  This in spite of the fact that they haven't installed new capacitors, VHF parasitic suppressors, step-start circuits, vacuum relays, electronic bias, direct grid grounding, aftermarket rectifier/metering boards - the list goes on and on.  

If you want a project and enjoy working on things, then go for it.  It's fun and there are a lot of ways you can make it uniquely your own.  If you just want to use the amp, my recommendation is to bring it up with a variac and use it until/unless something actually goes wrong with it.  It may outlast all of us if operated on a regular basis.

That's just my opinion.  I don't know as much as some of the other folks who hang out here so let's see what they have to say.

73,
Bob

You'll see a lot of arguments to change the bleeder resistors to 100K to reduce heat but if the originals haven't significantly changed in value, you can just reuse them.  The Heathkit engineers were pretty good at their jobs...

The new capacitors are definitely much smaller than the originals.  The technology has improved a lot since 1980.

Harbach sells a complete capacitor retrofit kit if you need one.

The best answer though may be to leave it alone unless something is actually wrong with it.

The factual answer is "no" but the practical answer is "sometimes".  While I agree of course that the surrounding circuitry affects the readings, you can usually get a feel for whether or not the transistor is shorted or open.

If your meter has a semiconductor test position (usually indicated with a diode symbol), test the B-C and B-E junctions.  In one direction the meter will usually give a quick beep and will show a reading of about .500 to .600.  In the other direction the meter will probably remain silent.  If you see a solid and repeatable indication like this, the transistor is probably OK.  Check a few transistors out of circuit first to get a feel for how your meter behaves, then give it a try with a in-circuit component.  The problem with this approach is that you need to interpret the readings and experience helps there.

The best way to test the transistor in circuit is to check the voltages with the equipment turned on (assuming you have steady hands).

http://www.worldradiolabs.com/download/Gonset_GSB-201.pdf

While keying, can you check the voltage drop across each component before and after the relay drops out?

Grainger has a wide selection of fans if you choose to retrofit.  I found a reasonable replacement for the blower in an old National in their catalog.

http://aaxeon.com/

You may have to call their support line and ask.   You may also have luck by searching for the part number on the integrated circuit instead of by the board model number.  Sometimes another vendor will use the same chipset and their drivers will work.