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

[Platefire]

I thought I had gone over all of my amps and had them working right but I just discovered it again hiding. I fired it up and it does have some noise problems. Does anybody remember GroundhogKen, Brad S Bryant and EL34(Doug) helped me with this in 2001?? They all pitched in and helped me get r dun. I really did a lot of playing over this amp but it's been sitting up for several years now. What little I've tried it, it has a humm problem. I changed most of the coupling caps way back then but didn't mess with the electrolytics.

So I tore it down today and checking the resistors and caps. Going to change all the filter caps. Making a list and checking it twice;>)I pretty much left the power amp as is and modded the preamp for guitar. This was a thrift shop find I got for $20. Its got 6SC7, 12AX7 pres, 12AX7 PI, PP 6L6's and a 5Y3 Rec.

[Latole]

Why not the complete schematic ?
Clean tubes's sockets contact and tubes contacts.
Clean pots

In both cases, good contact cleaner is a must ; I only use Deoxit

In an infrequently used tube amp, these contacts oxidize and can cause problems.


The tone stack as you show

[Latole]

Picture inside the amp may help.

[Platefire]

This was way before ExpressSCH was available or wasn't aware of it. I spent $32 and got the full set of documentations from SAMS Technical Publications on the amp. I may draw a proper schematic while working on it. Since it still has the 1959 filter caps in it, I'll change those first and see what that gets me. The amp was sounding pretty good as is but would start growling out of no where, stop and clear up. I've found some resistors that are pretty far off also. Yelp I use Deoxit.

I remember being really excited when I got this amp because it was in really good shape. So I had a really good platform and tube compliment to work with. There were several different Integrated hi-fi amps brands in this style that came out in the late 50's.This GE one is really built like a tank:>)

[bmccowan]

Radio Museum has clean copies of that schematic. I have one of those amps waiting for a soldering iron.
https://www.radiomuseum.org/r/general_el_pa_20.html

[Platefire]

I'm Thinking the best way to replace the filter caps is remove the old Filter Can, plug the hole with a cover plate and install axle filter caps on terminal strips. Here is a gut shot of the situation at the Can Caps. The old can is a Mallory 40/450, 40/450, 5/350, 100/50 and 1 3/8 Dia x 3" high. Not sure if any direct replacements are available but pretty sure it would be cheaper to do F & T Axle caps on terminal strips.

Also it has a umbilical cord from the power amp to preamp section where there is a 30-30/350V axle cap. I found an F & T 33-33/450V axle cap for that. Two + leads on one end and ground lead on the other.  It's also 1" Dia same as the old paper cap, so the new F & T will fit the existing holder bracket

I know this is a dangerous question but, After having a look, whatayathink?

[sluckey]

I suggest using a F&T 50/50 can to replace the two 40s. The can will fit the hole from the removed original can. Need the clamp too. Then use an 8uf@450V and 100uf@100V axials for the other two caps that were in the can. Plenty of room to install these small caps.

[Platefire]

I like that idea!!! Also Thanks for the schematic. I had marked up my old original so much that it's good to have a clean slate to look at.

I got a couple of questions I'm thinking on:

Discovered the NFB resistor is 3K and a metal film resistor.  The schematic shows 27K. I don't remember changing the resistor but nobody else could have done it because the amp was untouched when I got it. Well it sounded pretty grand with the 3K but I realize that's a lot of NFB going back to the circuit. So I'm thinking you would say experiment with the value or put a 50K pot it or put the 27K in there with a on/off switch on it?

I installed a standby switch on it originally but I had to extend the wires all the way out of the chassis to a switch on a bracket because of the chassis odd configuration. I did get bit once when my finger got on the wrong side of the switch. Is there a safe way I can make this work or should I just do away with it for safety reasons? I does have a 5Y3, so the B+ would come up slowly

[sluckey]

Put a 50K pot in series with your 3K resistor.

If you must have a standby switch put it inside a small enclosure. I suggest this Bud Box from Amazon, but to save $$$, you could even use a small PVC electrical box from Lowes.

[Platefire]

What about coating the inside connections of switch with Liquid Electrical Tape? As per Link:
https://www.amazon.com/Star-brite-Liquid-Electrical-Tape/dp/B0000AXNOD/ref=sr_1_3?keywords=Liquid+Electrical+Tape&qid=1686685364&sr=8-3
On the NFB, the problem with that is finding a place to put a pot. I'll have to study that one.
Maybe use a pot in series to find the best operation setting and then replace it with a fixed resistor.

[bmccowan]

If you must have a standby switch

That Sluckey comment is telling. Unless you intend to be gigging the amp and taking multiple liquid or dry refreshment breaks, I would see no need for the standby switch.
BTW - this is great. I often find that if I let some thrift shop treasure sit on a shelf long enough, someone else here will work on one and I get to lurk and learn.

[Platefire]

I can't really truly justify a standby switch other than I like um and it's important to me
 
The hardest part of these type integrated hi-fi amps like this one is the configuration of the chassis makes it harder to access everything. Once you get by that, it makes a nice head for what ever cabs you have. I use to have a old Peavey Musician 2-12 cab that I run this head through and it just sounded great and was a great match. My youngest Daughter run off with that to go with the Bogen CHB-50 Bassman conversion I made for her.  BTW, she was in the Army at the time, so I did it like this:

[Platefire]

Where does a fellow get a 7.5K 5watt power resistor these days. Guess that's an odd value? The existing one is measuring 6.3K

[sluckey]

Where does a fellow get a 7.5K 5watt power resistor these days. Guess that's an odd value? The existing one is measuring 6.3K

https://www.ebay.com/sch/i.html?_from=R40&_trksid=m570.l1313&_nkw=7500+ohm+5+watt+resistor&_sacat=0

[bmccowan]

I love that US Army Bogen. I bet it sounds great. I hope she still plays it. I have found Bogen, Stromberg Carlson, and Grommes to be really solid amps. As you say, the PA amps are easier projects than the integrated hi-fi amps.

[Platefire]

I have had the Bogen CHB-10, CHB-20(still have) and CHB -50. She was coming from Ft. Bragg to visit, pick it up and I was doing some last minute mods on it adding a master Vol. Moving some of the pots around and left the holes in the face plate. My intention was to install plastic plugs in the unused holes and put some text to identify the pre vol and master vol but she drug it off before I had time to finish. I made that amp when she was really into guitar playing. She had such great potential, fast learner but all of a sudden she dropped it and that was it. She bought an acoustic several years ago and bangs on that sometimes. She still lays claim(and has possession) to the Bogen anyway;>/

[Platefire]

sluckey Thanks, I bought that NOS Wirewound Sprague Koolohm 7.5K 5watt

Wow! I found a mistake in the power supply going to the pre-amp section. R30 in the power amp section feeding B+through umbilical cord to pre-amp is suppose to be 110K but with three orange bands is 33K??

I'm going to leave it as is for now. When I get it up and running, I will check voltages in
preamp and see if I need to change it out

[sluckey]

Wow! I found a mistake in the power supply going to the pre-amp section. R30 in the power amp section feeding B+through umbilical cord to pre-amp is suppose to be 110K but with three orange bands is 33K??

You probably did that on the advise of someone in order to increase the B+ for the preamp to more common guitar amp voltages. No need to change it if you are happy with the sound.

[Platefire]

yelp! Will wait and see the final results and go from there. Some of the voltage readings recorded way back then appear to be rather low for guitar amps, especially in the pre-amp section. Main thing right now is just identify and replace all bad components with new. I can tweak resistor or cap values after I get it up and running and see where I'm at.

[Platefire]

Outch! The more I look at this amp, the more I'm amazed that it worked as good as it has? Looking at the preamp today starting with the input jack. It appears that how you would normally hook up a 1 Meg to ground on the grid, I had a 15 Ohm:>))

Color bands Brown/Green/Black where the 1 Meg  Brown/Black/Green
I  must of got the color band arrangement confused and didn't bother to do a follow up check on the Ohms with a MM. Common greenhorn mistake. When I fix this thing right it will probably never sound the same again---but I hope it's for the better

Here is a link to some handy resistor color band charts:

http://www.bpesolutions.com/atechnical/resistorqv.pdf



 

[sluckey]

Outch! The more I look at this amp, the more I'm amazed that it worked as good as it has? Looking at the preamp today starting with the input jack. It appears that how you would normally hook up a 1 Meg to ground on the grid, I had a 15 Ohm:>))

In addition to the resistor mixup, there's another flaw in that mod. See attached pic...

[Platefire]

So both of those 3.3 Meg grid to ground resistors should remain?

[sluckey]

R13 is not causing any problems but it is not needed either. I'd clip it out.

[Platefire]

Well I got the input wired up as per your drawing for pretty near happy Grid leak with R13 gone to happy resistor grounds


I think I've looked it over enough to make a fairly intelligent parts order. There's always going to be something you missed no way out!

[Platefire]

Question? and I think I know the answer, but not sure? This amp on the 6SC7 Socket has the socket isolated from the chassis supported by two screws with nut on bottom side. The screws are surround by thick & tall rubber washers top and bottom, so the socket kind of floats on its own separate from the chassis.

I think it may be because the 6SC7 is known having micro-phonic tendencies and this cushioning helps to minimize that tendency but may be other reasons also. May have something to do with grounding, but not sure.

[sluckey]

I think it may be because the 6SC7 is known having micro-phonic tendencies and this cushioning helps to minimize that tendency but may be other reasons also.

Good guess. That was common practice for high gain preamps for mag cartridges that are very low output.

[PRR]

> 6SC7 is known having micro-phonic tendencies

ALL tubes are microphonic. (This was a claimed advantage for that newfangled solid-state junk.) Mounting on rubber was standard for the first audio tube, if the maker cared.

Guitar amps are *just* enuff less sensitive than mag-phono amps that you can usually omit the rubber. Or when hi-gain amps evolved, they had so many 12AX7 in them that you could swap-around and get a good tube at the input.

[Platefire]

After doing the changes to the pre amp section I couldn't resist hooking it to the power amp even though the main cover panels are still disassembled. While I was installing all the tubes and umbilical cord I deoxeted the tube sockets and tube pins. The pre amp worked good and I didn't notice any of that hummm breaking in and out as before, a lot quieter.

I also checked a few voltages. The voltages coming off the rectifier in the power amp under full load is running 17 to 23 volts over schematic values. With schematic values first then actual readings next, 1st node 425/448, 2nd node 325/354, 3rd node 315/332. After 332 runs through the 33K to umbilical cord, pre amp gets 252 with sch showing 255----so that's pretty close. Whoever resized the 110K down to 33K for the pre amp voltage knew what they were doing, I sure don't remember doing it!  The plate voltage to the 6SC7 is 66:>)

So the pre amp is ready to go except for the two 30/30 uF 450V cap I'll be ordering with the other filter caps to the power amp.

[sluckey]

What about coating the inside connections of switch with Liquid Electrical Tape? As per Link:

Don't put a band aid on a potentially dangerous shock hazard! Put that switch in a proper enclosure or just eliminate it. You don't really need a STBY switch. If you need a mute switch there are several ways to safely do that with no worries about shock hazard.

[Platefire]

Got a grounding question on installing my new can cap. I don't install these every day, so I'm not too sure of myself. Attached is a picture of the new can:

1-Don't I need to run a chassis ground wire to the negative terminal?
On the old can, (see picture) the PT center tap, 250 R41 Power tubes cathode Res, 25K R45 Grid 2 power supply ground and ground to umbilical cord are all grounded on old can ground term!

2-Can I just transfer all these grounds to the new can negative term?

Also I got a 100uf/100V for power tubes cathode bypass and a 10uf/500V individual capacitors to take the place of C1d and C1c formally provided by old can that will be removed.

3-I call your attention to the 4 term, 1 ground terminal strip at bottom of old can cap pix. My plan is to remove the old standby switch green and black wiring and install the replacement caps for C1c and C1d to that terminal strip and attach the old leads from the old can cap to that.

This is what I'm thinking, please tell me if it's Kosher. I'll also attach the schematic for your ref

[sluckey]

yes

[Platefire]

Thanks!

[Platefire]

Got a basic Tube Amp 101 Question? I know how I've done it in the past, but I want to know what would be considered the proper way to do it. That is-----Extending component leads when they are too short and you need to add a section on for it to reach.

Attached is a picture of my Pre Amp chassis showing the old 30/30uF /Cap in a metal clamp with long red leads and laying on top of that the blue F & T 33/33uF replacement.

So my intent is to clip the old red leads off the old Paper Mallory cap full length and use those to attach to the F & T. Of course I will then remove the old Mallory cap and insert the new F & T in the clamp and attach the new leads to the old Red leads.  Question is, what would be the best way to attach the leads, solder it and finally insulate it?

[bmccowan]

I have replaced a lot of filter caps in old Gibson amps. Many of them have that exact setup - dual firecracker caps. Those F&T dual axial caps are perfect for that. I don't know if there is one "proper" way, but what I do is clip the leads on the F&T down to about an inch or so and make an eyelet. I slide a short length of heat shrink over the old cap leads, make a solid solder connection and then insulate with the heat shrink. I've tried twisting the leads and soldering them, but the F&T leads are stout, solid wire and the firecracker leads are stranded and flexible - I have a hard time making a good twisted connection with the two different wire types.

[Platefire]

Thanks very much for your reply. In the past, I have twisted the leads together in a overlap of about an inch. It never worked well and was hard to do. I did a little research and found a pretty common practice was to bend a small hook in each lead, mechanically tighten the hook with pliers and then solder--kinda like the attached picture.

It hadn't occurred to me that the leads on my old firecracker might be stranded. I checked and thank goodness it isn't. So unless someone enlightens me about a better way, I think I'll try the hook method.

We do these things in our builds/repairs but never talk about it. Rather than do what I normally do, I just wanted to maybe find a better way I could feel good about after its all said and done

[sluckey]

The hook is a good method. However, I would cut the new cap lead about 1/2" long and put the hook as close to the cap body as possible. Then slip a short piece of heatshrink tube over the old wire before making the solder connection at the cap body. Finally, slip the heatshrink over the solder joint and apply some heat.

[Platefire]

Got Ya! Thanks sluckey

[Platefire]

Well I've been busy on this last couple of days. Got all the new filter caps in. It's sounding much better and running nice and quiet.I'm now checking a lot of the old carbon resistors to see if they are holding or drifted. On V2 the driver and recovery for the Baxandall tone stack has resistors that I've measured in place that's is reading pretty off:

V2A R25 680 Blue/gray/Brown Pin #3 Cathode Resistor to ground reads 853 measured in place

V2B R26 470K Yellow/Violet/Yellow Pin 7 Grid to Ground reads 90K measured in place

V2B R28 1800 Brown/Gray/Red Pin 8 Cathode Resistor to Ground reads 2100 measured in place

Don't know if these are off enough to sweat blood over? Problem is these resistors are what I would say, almost inaccessible because of the "V" shaped pre amp chassis. To get your hands plus a soldering iron in that space??? See picture below of the chassis.
Also here is another top view of the chassis and right below the pilot light is V2 socket with subject resistors. Maybe you know some tricks to pull such a task off. Won't really know how far off they are unless I get one leg disconnected and measure it that way
 

[bmccowan]

Hmm - those pics are curbing my enthusiasm for working on mine. Any chance of forcing that bend open enough to get in there? Anyways, R25 and R28 I suggest leaving as is for now. As bypass resistors, those values might work out fine. R26 - I wonder if a parallel path is affecting that reading? Does not seem likely that it went from 470K to 90K.
Edit - although the ground side of that 470K does not look to too tough to reach - I think I would make notes and continue on to the point where you can sound check and then reassess.
BTW I have a Hako desoldering tool with a built in vacuum - they are pricey but are great for such work.

[sluckey]

Here's a tip when measuring resistors in circuit...

If the resistance is higher than expected, the resistor has likely drifted up in value. If the resistor reads lower than expected there is usually something in parallel to the resistors. Resistors rarely if ever drift down in value. If the resistor shows signs of being overheated the value could measure either higher or lower.

Always have a schematic handy when checking values. Cathode and plate resistors can usually be measured accurately in circuit. Consult the schematic for most others to see if there is something in parallel with the resistor. It's usually apparent.

[Platefire]

Attached is the schematic for the input and tone stack stages I made 22 years ago. It's pretty much correct except for the Grid Leak changes recently done on the input. Also R13 has been removed. I don't see anything I would consider in parallel with those 3 resistors, do you?

[sluckey]

According to the schematic you should be able to measure R26 accurately in circuit. It's highly unlikely that a 470K has drifted down to 90K. I would recheck R26 and verify the amp is actually wired according to the schematic.

Does the amp really have that .1µF cap just prior to V2 pin 7? If not, that could explain the 90K reading.

[sluckey]

Hmm, after comparing your schematic to your pic, I'm gonna say your schematic is WRONG! I see a dark green wire that appears to go from pin 7 directly to the junction of a 100K and a small ceramic picoFarad cap. No .1µF cap in sight.

[Platefire]

Thanks sluckey
The .1 uf just prior to V2 pin 7 does not exist. The dark green wire from pin 7 goes directly to center output term of the treble pot. The 100K coming from the bass pot is correct.

The small 330pf ceramic cap is being fed from V2A pin 1 through .1 coupling cap through the ceramic to input term of treble pot, just like the schematic. So even though the 330pf appears to be connected to the 100k your seeing, it's not. However the 330pfis connected to a 100K underneath the terminal strip that is connected  to the input term of the bass pot.

I hope this clears thing up a bit. I can't believe the 470K on pin 7 could have drifted down to 90K either. Something is effecting that reading!
I did start a new ExpressSCH schematic on this amp but haven't been able to make much progress lately. I know a good schematic would clear a lot of things up and help identify all my original mistakes but we will git r dun

[sluckey]

The .1 uf just prior to V2 pin 7 does not exist. The dark green wire from pin 7 goes directly to center output term of the treble pot. The 100K coming from the bass pot is correct.

There you go. That's the parallel path. The blue path is parallel to the red path. With the Bass pot turned down you have a 100K + 10K = 110K parallel to the 470K. That calculates to 89K. With the Bass pot turned up you have 100K +1M + 10K = 1.11M parallel to the 470K. That calculates to 330K.

Measure the 470K again but turn the Bass pot. You should see the meter reading change.

[Platefire]

bmccowan
I've got an old Weller 8200 N that's literally held together with electrical tape. I use that for everything these days. I've tried the simple 30 watters and have a bunch of them I never use. They just didn't work for me. That Hako sounds great with a vacuum to suck that solder down. I wouldn't know how to act with one of these modern total solder stations.

On the old amp, knowing what I know today, about the difficult configuration, I would have probably sold the amp or parted it out and used the parts to build something else. It was in like brand new condition when I got it and I was so excited because I though it was so Cool. So I dove right into it not giving it a second thought. The crazy thing about it, is spite me being a total greenhorn and had no idea what I was doing, the Hoffman forum walked me right through it and it ended up sounding great to my ears. So I've used it quite a bit over the years but lately it was starting to moan and groan. So I'm a little cautious about what I change on it because that tone and feel I've enjoyed may be part of the mistakes I made---I don't know. However on the other hand, it could be improved because it does have some quirks that I've never flushed out. So it hard to know how far to go with it. I've been running some of my different guitars through it and it seems to be friendly to every guitar. Just amazed at the tone and feel. Platefire

[Platefire]

With the bass pot full up I'm reading 372K across the 470K resistor. That's about 42K more than you calculated. Does that mean my 470K is salvageable? Those kind of calculations are really valuable, don't see how you retain so much. Mine seems to leak out

[sluckey]

With the bass pot full up I'm reading 372K across the 470K resistor. That's about 42K more than you calculated. Does that mean my 470K is salvageable? Those kind of calculations are really valuable, don't see how you retain so much. Mine seems to leak out

What that 372K reading means is your bass pot is really more than 1M. The 90K reading you had with the bass pot turned down (and out of the circuit) means your 470K is perfect. I calculated 89K and you measured 90K. I bet my calculator is more accurate than your meter.   

The parallel resistor equation is easy... It's the reciprocal of the sum of the reciprocals. Ain't that easy? 

[Platefire]

Great and Thanks! If I put a new battery in my MM, it would probably read up to snuff

What about R25(680/853) and 28(1800/2100). I think I could access R28 fairly well with extreme caution but R25 is a pure bugger

[bmccowan]

bmccowan
I've got an old Weller 8200 N that's literally held together with electrical tape. I use that for everything these days. I've tried the simple 30 watters and have a bunch of them I never use. They just didn't work for me. That Hako sounds great with a vacuum to suck that solder down. I wouldn't know how to act with one of these modern total solder stations.

On the old amp, knowing what I know today, about the difficult configuration, I would have probably sold the amp or parted it out and used the parts to build something else. It was in like brand new condition when I got it and I was so excited because I though it was so Cool. So I dove right into it not giving it a second thought. The crazy thing about it, is spite me being a total greenhorn and had no idea what I was doing, the Hoffman forum walked me right through it and it ended up sounding great to my ears. So I've used it quite a bit over the years but lately it was starting to moan and groan. So I'm a little cautious about what I change on it because that tone and feel I've enjoyed may be part of the mistakes I made---I don't know. However on the other hand, it could be improved because it does have some quirks that I've never flushed out. So it hard to know how far to go with it. I've been running some of my different guitars through it and it seems to be friendly to every guitar. Just amazed at the tone and feel. Platefire

Weller 8200N - Is that one of those big pistol grip things? Oh man!
The Hakko desoldering iron - F-301 is also pistol grip, but its really only good for desoldering. Tip is too bulky for soldering. I have another Hakko station for that. I have a modern Weller station too, but the damn thing has a mind of its own. Sometimes powering down to 350 from 750 - other times staying right on the setting. I should have returned it long ago.
The amp - I forgot that you previously had it operating to your satisfaction. Mine I have yet to touch, and may resort to putting the trannies and a few other parts in a new chassis. That folded chassis looks like a nightmare.