My First Amp Conversion 22 Years ago, General Electric PA-20 Mono Hi-Fi

[sluckey]

Good to see you using your noodle to trace the parallel path and calculate that resistor puzzle. Now you need to spend some time on the resistor color code. The color code for 27K ain't red/blue/orange.   

[Platefire]

Well OK, Red/Violet/Orange 

It looked blue on my Radio Shack, 3 wheel, Resistor/Capacitor Color Code Tool

[PRR]

> It looked blue on my Radio Shack, 3 wheel, Resistor/Capacitor Color Code Tool

That's been out in the sun too long.

I suppose your Shack shut-down??

You can find them, but shipping is silly.
https://www.tequipment.net/ElencoCC-100.html
https://vetco.net/products/resistor-color-code-calculator-card
Table on back of their business card, free for the asking:
https://www.allspectrum.com/store/resistor-color-code-chart-all-spectrum-electronics.php
Stickers:
https://www.redbubble.com/shop/resistor+stickers
T-shirt:
https://www.redbubble.com/i/t-shirt/Resistor-Color-Chart-by-Cheesy-Puffs/25352665.WFLAH
On-line cheat:
https://www.digikey.com/en/resources/conversion-calculators/conversion-calculator-resistor-color-code

Print yourself:
https://www.electronics-tutorials.ws/resistor/resistor-colour-code-wheel.html
https://www.electronics-tutorials.ws/wp-content/uploads/2018/05/pdf-colourwheel.pdf

[PRR]

me> shipping is silly.

https://www.tequipment.net/ElencoCC-100.html
Product Total $1.90
Shipping Estimate

• Ground - $11.92
• 3 Day - $32.67
• 2nd Day Air - $40.10
• Next Day Air - $89.56
• USPS - $12.95

Total  $13.82

[sluckey]

Maybe one of the Navy boys will chime in and tell you how real techs memorized the resistor color code. My trade school instructor was retired navy so I got it a long time ago.

[PRR]

Wikipedia has an AMAZING collection. Most of them gibberish, and apt to confuse black and brown.(*) (There are three "b" colors.)
https://en.wikipedia.org/wiki/List_of_electronic_color_code_mnemonics

(*)My trick, beyond offensive sayings, is to KNOW that 100 Ohms, a super-common value, IS Brown-Black-Brown.

[sluckey]

The one I learned made the top of the list in the Offensive/outdated section.

[Platefire]

Wow! That's a lot of Resistor color code help there, Thanks! PRR. I actually started trying to memorize number/colors combinations tonight. Got a good start on them. Seems to me if you get the color/number combinations memorized plus the multipliers, you've got a good start on getting correct values. If you can't calculate it in your head, using your calculator on your phone, you can crunch the number pretty quick. I'm talking about 4 band color codes, not 5 or 6---not there yet

[sluckey]

There are many apps for resistor color codes for android or apple.

I've been using the 4 band color code all my life. It's ingrained in my brain just like the alphabet. I recognize resistor values without having to think about it. I'm a bit slower with the 5 band color code and do have to think about it. Since I changed over to metal film I've gotten quicker. As long as the third band is always black (0) I'm pretty quick. But as soon as the third band changes color (such as 472K) I have to make a deliberate calculation in my head. Thankfully, I don't own any of those weird values.   

[bmccowan]

The one I learned made the top of the list in the Offensive/outdated section.

When you said Navy, I knew that was the one. Not safe anywhere anymore.
I like that one from Canada: Black Bears Roam Our Yukon Grounds But Vanish in Gray Winter
Although I find it strange the way the mind works. I tend to just recognized the value with some kind of photographic memory - but I have the hardest time with the multiplier.

[sluckey]

The multiplier is easy if you don't think of it as a multiplier. I think of it as the number of zeros to add after the first two digits. For example if the third color is orange I just add three zeros. For example, yel-violet-orange is 47 with three zeros, or 47000. Or, red-red-black is 22 with no zeros.

[bmccowan]

That should help me. I have a Nephew who is both color blind and dyslexic. He would love this hobby.

[PRR]

The multiplier is easy if you sort resistors by multiplier color. I had a black bin, a brown bin, red, orange.....
If I needed "about 3k", look in the red bin, find a red-vio-red (2.7k) or org-org-red (3.3k), whichever came to light first.
When I needed 2.2Meg I learned to look in the green bin. So only 8 bins, 7 if you admit that the blue-stripe parts can sit in the green bin (IF you have good blue/green light and vision).

[Platefire]

Got the metal film resistors for this project in today. Weird, the ticket said for $2.99 you get 10 pieces for each R value. They were actually 17 each. Not complaining:>) Also I got a little concerned because one package was marked 2K, but I didn't order 2K but 3K. No worry when I checked it with my MM, it was 3K.

Installed the needed resistors today, about 11. Fired it up with a light bulb limiter, no problem! When put it on household current that voltages are running 10 to 45V higher than schematic. I jammed on it this evening and it's sounding really good. No red plating. No unwanted noises

[Platefire]

Attached is my draft ExpressSCH. that was never done back in 2001. So I welcome you to check me out! I plan to add some more info as I get time, but this is the basic draft.

It calls for a 5AMP fuse. That seems a little stiff to me.  Seems like a 2 AMP would be more in line to provide some real protection?

The voltages shown are recent readings.

Like to note a difference in what the original sch shows and what's actually been in
place. R30 shows a 120K but there is actually a 22k in that place. Also R47 snows a
22K and there is actually a 120K in that place. When I installed new resistors, I replaced
to match what's in place rather than the schematic.

FYI-The main changes has been to the Preamp section. The power amp is stock like the original schematic. I have attached the original schematic for reference. Platefire

[bmccowan]

Congrats on bringing her back. Tough chassis to work on! And thanks for posting a draft completed schematic.

[Platefire]

Thanks! I think it's a lot of sentimental value with this one. Of all the hunting I use to do for old tube stuff, coming up empty handed most of the time, this old amp I picked up in like new condition for $20 really boosted my desire to continue the hunt.There is a lot of those old PA-20's still floating around, maybe someone will use my plan to re-build one. But you really have to be patient with this one, because it's an extra level or two more of difficulty.

On ExpressSCH, I haven't used it in a while and was working on this amp sch. Spent the better part of 3 hours working on it and all of a sudden the program shut down and disappeared. So about that time I remembered that I always used to be hitting the "save" disk symbol because this program, as far as I know, has no automatic save. Sure enough I opened the file back up and lost everything. So when I'm working on it, I'm now hitting save every minute of so!

[Platefire]

Looking back at this amp after latest restoration, I don't understand the power nodes routing. Usually the second power node(B)goes to the power tube screens but on this amp, it goes to the PI. Then the 3rd power node goes to the screens. This seems to be going against the norm--any engineering rational behind this?

[tubeswell]

Looking back at this amp after latest restoration, I don't understand the power nodes routing. Usually the second power node(B)goes to the power tube screens but on this amp, it goes to the PI. Then the 3rd power node goes to the screens. This seems to be going against the norm--any engineering rational behind this?

You mean B and C are split rail (from the same B+ node) but with C feeding the screen (and C is lower than B). Probably to keep the screens sitting well below the plates, likely to do with having cathode-biased 6L6s idling at near Pmax and I suspect the load line would otherwise cross the Vg0 grid curve below the knee if the screen voltage was the same as the B+ (so having a lower screen voltage lowers the grid curves to keep the load line through the knee - helping to keep the screens happy when driving the amp hard*). No doubt these were over-designed to ensure longevity.


*screen current tends to shoot up when the plate voltage swing bottoms out ('below' the point where the grid curves all drop suddenly to the origin). If the load line  crosses below the Vg0 knee, then the probability of over-dissipation of screens is higher when plate voltage swing bottoms out. Lowering Vg2 is one way to avoid this. Biasing the output tubes colder is another way. Running a lower reflected load resistance is another way.

[PRR]

Old 6L6 had low Vg2 ratings; also we though we had enough power at 300V on G2. But it had to be non-saggy, hence the divider with heavy bleeder.

But also: you are looking at a 2023 interpretation of a 1959 design. Might want to look at how it used to be. Lower voltages, not very different; but one feed much cleaner.

[Platefire]

OK, thanks for taking a look and the explanations. I'm use to the screen being just a few volts below plate, not use to that much difference. I was getting ready to put the amp back together and wanted to make sure everything is OK before hand.

The original tubes were originally 6L6GB's and I've been running Sovtek 6L6WXT's last 22 years, so I don't guess I've overloaded my wagon, it's still kicking and no burn outs so far.
That lower screen grid voltage may have been my saving Grace

[Platefire]

sluckey---I did remove original K2 component off the master volume pot. Made quite a difference. Before I was having to cut bass fully counter clockwise and boost treble fully clockwise to get the tone I wanted.

Now---The amp is brighter and the tone controls works more like they are suppose to with adjustments landing closer to the 12:00 position. I guess I was kind of afraid to remove that because I wondered if it would remove some of the original charm, but it only improved the amp. So that old myth is abolished. Thanks!

[PRR]

I'm use to the screen being just a few volts below plate, ....originally 6L6GB's

6L6GB specs: https://frank.pocnet.net/sheets/127/6/6L6GB.pdf

400V Vp, 300V G2

The GE runs 415V-25V= 390V on plate-cathode, 315V-25V= 290V G2 to cathode. Pretty much all they dared.

[Platefire]

Both transformers have a 1957 date, so it may be a year so older than I had it dated(1959). I put it all back together and been testing it out with different guitars.

Inside the top cover panel right above the power tubes, it had a 1/8" layer of asbestos about 4" x 4".  I guess that was to protect the panel from the heat. The power tube are only about 1/2" below the cover. I carefully removed it. I'm pretty familiar with that, we did a lot of asbestos abatement contracts at Fort Polk. Never seen it in a tube amp before. I've been reaching over to feel that spot on the panel when I using the amp but it only gets warm, not hot. I might ought to replace that with some modern/safe heat resistant material. If anybody has one of these amps, you need to be aware of that.

[bmccowan]

If anybody has one of these amps, you need to be aware of that.

Thanks Platefire - As you know I have one of those GEs sitting on my "maybe one day" shelf. But my intent is to use the iron and a few other parts and build into a non-nightmare chassis. But when taking it apart I will watch for that patch and be careful. My understanding is that you want to avoid the asbestos becoming friable. It's inhaling the fibers that is not so good for you, eh?

[Platefire]

Yes, I remember you saying you have one. Believe me, your saving yourself a lot of problems by putting the trannys in a more conventional chassis. When you tear it down for parts just be careful removing that top cover not to disturb it and put the whole top panel in a sealed HD bag and dispose of it in the proper place. 

[Platefire]

Thinking about something? The asbestos heat shield over the power tubes is removed, but I noticed last night the cover plate was getting pretty hot, especially on the under side. Guess they had it there for a reason:>) I've been looking a self adhesive heat shield material that protects up to 1000 degrees to put on there but I'm kind of concerned  about the self adhesive letting go after so long. The attached picture shows the situation, the small space above the tubes between the top panel, about 1/2". Playing outside on a hot day it could really really get hot.

Any recommendations for this or am I making a mountain out of a mole hill. 
 

[bmccowan]

Drill some ventilation holes in the case?

[sluckey]

It's a metal case and the back is completely open. I'd let it be. Really, how often will you be "Playing outside on a hot day"?

[PRR]

... how often will you be "Playing outside on a hot day"?

And I love to live so pleasantly
Live this life of luxury
Lazin' on a sunny afternoon
In the summertime
In the summertime
In the summertime

As a 1957 Hi-Fi you might build it into pine cabinetry and leave it on all the time. Or your child might touch the top and get a blister.

Live dangerously!! (And smarter than the average bear.)

[sluckey]

Lazin' on a sunny afternoon
In the summertime

These days it's more like...

Hot town, summer in the city
Back of my neck getting dirty and gritty

[bmccowan]

Where you live - yes.
In Maine its more like:
If the rain comes
They run and hide their heads
They might as well be dead
If the rain comes

Either way, I like your solutions better. With Sluckey's you need do nothing. With PRR's you just need to mix up a gin and tonic and Beware the Demon Bowler.

[Platefire]

Wow! I've had one crazy day The good part is I get home and find ya'll have my problem solved and as it turned out, I didn't have a problem at all.

My Amp, My Amp                                                                                   
has got me thinking bad thoughts
My Amp, My Amp
Could of been  cooler and got mellowed out
but keeps picking at me, oh what shall I do?
Nothing said I, Nothing is what I'll do

[Platefire]

Made A Cover for it