Silvertone 1484 - Help me understand the power supply

[LooseChange]

See the schematic below. I am confused with the power supply. Specifically the connections from the Power Transformer to the B+.  They use two sets of windings. Can somone give me an idea of how this works.? Thanks!

[sluckey]

It's two voltage doublers, stacked up in series aiding. Just like two flashlight batteries.

[Fresh_Start]

It's two voltage doublers, stacked up in series aiding. Just like two flashlight batteries.

OK, I sort of get that but why is the choke connected to the junction of two caps instead of the same point connected to the OT center tap?

Why aren't smoothing resistors needed in parallel with each of those 100uf/150v caps?

Chip

[sluckey]

why is the choke connected to the junction of two caps instead of the same point connected to the OT center tap?

Because they wanted a lower B+ voltage feeding the screens and other circuits. That's just an easy (and cheap) way to get a lower voltage without using a big dropping resistor.

Why aren't smoothing resistors needed in parallel with each of those 100uf/150v caps?

Do you mean equalizing resistors? They're not needed. Just consider the top voltage doubler (D1, D2, C26, C27). D1 pumps up C26 and D2 pumps up C27 equally. Just the nature of the circuit insures that the voltages will be the same (assuming perfect diodes and caps). Since the caps are connected in series aiding, the voltage across both caps will be doubled (ie, V_c26 + V_c27).

Now really get creative and build another identical doubler circuit and stack it up in series with the first doubler. Just like batteries in a flashlight, the total voltage is the sum of each individual battery (err, doubler circuit).

Looking at the resistance readings for the secondaries, I'd say that the bottom doubler puts out more voltage than the top doubler. No matter. The total voltage is still the sum of all 4 caps.

Sears was notorious for building cheap electronics in the '60s. I suspect they saved a bunch by using those 150v caps in a 480v supply. This power supply fits right in with the cheesy reverb circuit and paper covered press board cab.

[Fresh_Start]

Sears was notorious for building cheap electronics in the '60s. I suspect they saved a bunch by using those 150v caps in a 480v supply. This power supply fits right in with the cheesy reverb circuit and paper covered press board cab.

If you applied this approach to this transformer with two 115 volt secondaries, would that give you an un-loaded B+ of 644? [(4 x 115) * 1.4]

Would using the voltage doublers cut the current rating in half as well?

Just curious...

Chip

[Megachunk]

On a non-tech side note, I put an adjustable bias in one, and it's a singing little amp! It sounded super sweet through a newer cab. What are you going to do with it, LC?

[Fresh_Start]

Sorry if I'm not adding to LC's original question, but this amp has a really unusual approach to deriving the signal for the reverb driver and then returning it to mix with the clean signal.  I can't quite wrap my brain around what is happening at the junction of R25, R26 & R27.  I think that the clean signal is being attenuated, relative to the reverb signal, by R25 & R27.  If that's really a 560K plate resistor (R26), it's the biggest one I've seen yet for a 12AX7.  Is my guess that the big value is needed to isolate the reverb signal from the clean signal close to correct? 

This is kind of like the 6G16 Vibroverb in terms of splitting a plate resistor to attenuate the clean signal, but there's an extra triode in the clean signal path.

LC - sorry for cluttering up this thread.  It's just a neat circuit to work through.  Shorting the grids of the power tubes in a push-pull amp will mute it, but I'm not sure "standby" is the label I'd put on the switch.

Chip

[jjasilli]

Fresh_Start, you're raising good points.  I've been grappling with voltage doubling circuits.  They're not so common in modern amps and info is hard to come by.

OK, I sort of get that but why is the choke connected to the junction of two caps instead of the same point connected to the OT center tap?  Yes, per sluckey, that's one way top tap-off of a lower voltage, but I suspect it's got the downside of only half-wave rectification.  Maybe that contributes to the "charm" of this amp.

Why aren't smoothing resistors needed in parallel with each of those 100uf/150v caps? I've seen such a schematic once, but for a doubling circuit off one HV secondary.  Because this design uses "stacked" HV secondaries, bleeder, or balancing, resistors might cause trouble.  Also, the doubling circuit, by its inherent design, already distributes voltage between the series caps.  So bleeder resistors may be superfluous.  Most grand old designs do not use them.

If you applied this approach to this transformer with two 115 volt secondaries, would that give you an un-loaded B+ of 644? [(4 x 115) * 1.4]  Yes.  And you don 't need more than one HV secondary for the voltage aspect.  You can keep piling up doubling circuits off of one secondary, but at the expense of: a) decreased filtering, and b) increased current draw.

Would using the voltage doublers cut the current rating in half as well? Yes.  That's no doubt why Silvertone used 2 windings, to distribute the extra current draw among 2 windings.  

Note that the caps are rated for only 150V.  In the old days extra tranny windings were cheaper than hi voltage filter caps & diodes, which may not even have existed in the ratings required.  Hence the existence of voltage doubling circuits back then, and their demise with modern hi voltage caps & diodes.

[HotBluePlates]

Why aren't smoothing resistors needed in parallel with each of those 100uf/150v caps?

What sluckey said, plus the added bleeders would cause problems.

Look at the top secondary winding, connected to D1, D2, C26, and C27. There is an upper and lower wire coming out of the secondary of the drawing.

Imagine for a moment that the voltage on the secondary is such that the upper wire is momentarily positive and the lower wire is momentarily negative. The lower wire being negative reverse-biases D1, so no current flows through it at this moment. However, D2 is forward-biased, so current flows through the winding and D2 and charges C27 to the peak voltage of the winding.

Now imagine the secondary voltage polarity switches. The upper wire is now negative and the lower wire is positive. D2 is reverse-biased and no current flows through it. D1 is now forward-biased, so current flows through the winding and D1, and charges C26 to the peak voltage of the winding.

Because D2 was reverse-biased and there was no current path for the cap to discharge, C27 still has the full peak voltage of the winding across it, and now C26 also has the full peak voltage of the winding across it. So the voltage on the caps adds and you have Vrms * 1.414 * 2 across the pair of caps.

But that is true if no current gets pulled from the supply, just as our B+ in any other amp rises up to the peak voltage of you pull out the output tubes. So really, the amp will draw off some current from these caps and reduce the supply voltage somewhat. Now, if you were to throw bleeder resistors across the caps, they would need to be quite large values, because they would tend to drain the lower cap while the upper cap is charging and vice versa. You could use them, but it would tend to knock down the output voltage and possibly the current capability of the doubler supply. And doublers (and higher voltage-multiplier circuits) are already known to have lower current capability than full-wave supplies.

this amp has a really unusual approach to deriving the signal for the reverb driver and then returning it to mix with the clean signal.  I can't quite wrap my brain around what is happening at the junction of R25, R26 & R27.  I think that the clean signal is being attenuated, relative to the reverb signal, by R25 & R27.  If that's really a 560K plate resistor (R26), it's the biggest one I've seen yet for a 12AX7.  Is my guess that the big value is needed to isolate the reverb signal from the clean signal close to correct?

I'm not really familiar with this circuit, so I'd invite corrections. But I think it's not to isolate the clean and reverb, but to mix them.

In a Fender amp, the same point is used to split off and return the reverb signal. So the big 3.3M resistor diverts most of the signal down into the reverb circuit.

Here, the reverb is split off at a triode grid, and returned at its plate. So the tube does the isolating for you.

Forget R25 for a moment. Both Channel 1 and channel 2 have 68k plate resistors fed by a common 220k resistor, which then goes to the power supply. Yes this is a tapped plate load like the Fender brown circuit, and the total plate load for each channel looks like about 290k. But because they share the 220k resistor, each channel's output is mixed as they're feeding C12. R25 is taking this mixing to the next level by mixing in the reverb signal along with the dry signals that feed the coupling cap.

I'm not 100% on this, but it looks to me like R27 is being used as a common impedance to mix all 3 signals. That also means the reevrb recovery's plate load is about 800k! That would give the stage a lot of gain, but very little current variation. If I'm right about the common impedance mixing, then it's the small current variation that's important. That would keep the reverb signal from drowning out the dry signals. I note that there is not a Reverb pot in the recovery circuit (like Fender) but in the drive circuit. So it looks like the idea is to never allow the reverb to completely swamp the dry signals, even if the reverb drive (Reverb Depth) pot is full up.

That's my guess anyway...

[Fresh_Start]

HBP - I believe we can refer to that as an "Educated Guess"

Thanks!

Chip

[PRR]

They are 400V rectifiers.

The amp wants 480V.

Using the "conventional" CT winding and two diodes, we need 960V diodes.

Using a single winding and a doubler, we need 480V diodes.

Splitting to two windings and two doublers, we only need 240V diodes. This is reasonable safety margin on 400V parts.

It also gives a 3/4 tap.... 480V on G2 is reckless, this gives 360V without much sag or heat.

There's also the bazillion AC/DC radios eating two bazillion 150V caps, making 150V the cheapest cap of the day.

The 6.2 5.75 ohm winding DCR difference is probably insignificant. Probably the same number of turns, same voltage, but one is wound over the other, longer mean length of turn.

> cut the current rating in half as well?

That tranny is rated 600VA. It can support 300W-400W DC. Do you have an amp which needs that much DC? (I can think of a few....)

But unless you need a 3/4 tap, or can't get high-volt diodes, I would (and have) wired the 120+120 as one 240V winding, double that. You get 240V*1.414= 330V each side, 660V total, more or less. You need >330V cap rating; 400V snap-caps are cheap today.

> what is happening at the junction of R25, R26 & R27.

Yes, very odd, isn't it?

Of course it is a mixer, a "plate mixer".

The 220K is a DC plate load for three tubes. Just sticking plates together on a common resistor "works", but distorts easy. The 68K in the dry-channel plates helps that a lot. The reverb tank is Piezo, apparently VERY high output, because after one stage of AX7 they use 560K to waste signal.

[RicharD]

Here's a good link about voltage multipliers:
http://www.allaboutcircuits.com/vol_3/chpt_3/8.html

99 times out of 100, voltage doublers are used for economic reasons.  The only "hi end" circuit that I can think of that used a doubler is the Altec 1567 and I'm not really sure I'd call that hi end.  Just about every single tube console stereo Sears & Robuck Co. ever made used a voltage doubler.  Voltage doublers do produce full wave rectification but the trade off is current.  Each time you double the voltage, you half the current.  That could quickly become a problem in a high power amplifier.

[jjasilli]

Buttery:  Voltage doublers do produce full wave rectification  At the top where the voltage is doubled, there is full wave rectification.  But if you tap-off for 1/2 voltage in between the 2 series filter caps, I think you have only 1/2 wave rectification. 

But, then again that's for a "standard", one secondary HV winding.  I'm thinking that here the choke is tapping off about 3/4 of the full voltage (the 2 windings are off a bit, probably to help identify which is which).  So it looks like the bottom winding-doubler circuit is contributing full wave rectification to the top half, and then to the choke; and the top winding-doubler circuit is contributing 1/2 wave rectification to the choke, and full wave rectification to the plates.  If so this is imperfect.  But imperfection is the glory of guitar amps.

Just saw the play the Addams Family:
Gomez Addams to his wife Morticia:  Are you unhappy, my Dear?
Morticia to Gomez:  Yes, Darling, completely!

[LooseChange]

Thanks guys!
I'm going to pop the hood this weekend.  The amp let some smoke out.
The Power Transformer is tiny! It's footprint is smaller than a Champ PT but it is about 2" tall (laydown type).

[Fresh_Start]

Thanks guys!
I'm going to pop the hood this weekend.  The amp let some smoke out.
The Power Transformer is tiny! It's footprint is smaller than a Champ PT but it is about 2" tall (laydown type).

Pics!  We need pics of the insides of this beastie!

Seriously, if you have time please take a couple of photos of the guts.  Given some of the odd aspects of the schematic, I can only imagine what the layout is like.

Cheers,

Chip

[HotBluePlates]

Pure rat's nest fun, if it's like any of the Valco/Supro's I've seen. Terminal strips with component leads wrapped around, but mostly parts running straight from one tube socket to another.

They can be a bear to trace out the schematic on your own.

[SoundmasterG]

Here are a couple pics of a stock 1484 Silvertone.....they were made by Danelectro btw.

Greg

[SoundmasterG]

And here are a couple pics of the one I modified for my brother also.....as you can see...completely different and a lot more going on......

Greg

P.S. Sorry for the poor picture quality....I need to take some new pics of the thing......

[LooseChange]

Mine looks exactly like the stock version. Still shiny inside.
Butter... What kind of mods???

The reverb feeds back when you turn it up.

[SoundmasterG]

If you do a search here you can find some threads where I talked about what I did to my brother's amp in detail, but in brief, I reordered the preamp tube layout to make sure the signal wouldn't cross over itself, added Fender reverb circuit and tank, built a new head box for it, added an EF86 tube and otherwise modified channel 1 for more gain, modified channel 2 and revoiced the tone stack due to increased gain, added an adjustable bias with a voltage doubler to get adequate adjustment range, put the trem and reverb on both channels, moved the screen voltage tap to the same wind as the plate and added a decoupling stage which increased power to 30W RMS instead of the stock 25, positive DC biased the filament center tap, optimized the grounding, replaced every part in the amp....thats about it. Took awhile. :)

I would guess on your reverb problem that there is too much gain from that stage, or perhaps the reverb tank itself is loose inside....My brother's never had the stock reverb tank so I never messed with it. I do have a completely stock 1484 here that has had the electrolytics replaced but has an oscillation issue. It does function, and the reverb works perfectly....but I don't have the time to further fix it or look at it right now. I hope you're able to fix the issues your amp is having....


Greg

P.S. Forgot to mention that the power transformer is the big one in the center of the amp (worst possible place for it) and the output transformer is the little tiny one over by the power tubes. That small output transformer is a large reason why the amp is only 25 watts stock, as with those voltages it should be more like 40 watts to 50 watts. Those small output transformers often blow even with the stock amount of power.

[LooseChange]

Thanks Greg. Now that I look inside I realize that the OT is the smaller one.
So here's the problem....
The amp is blowing fuses. I pulled all tubes, Disconnected all HT wires (4) and the bias supply wire. Still blowing fuses.  (Actually using a light bulb and it's going full bright).
Any thoughtsas to what the problem is... Hope it's not the PT.
Thanks!

[sluckey]

Not much left. Disconnect the filaments to see if that's it. Then disconnect the primary to see if that's it.

[SoundmasterG]

The PT is probably bad....it happens with these often. I'm not sure if Weber does a replacement for these or not....I believe Mercury Magnetics does, but they are spendy. You can also look on ebay.....sometimes you get lucky and can get a carcass amp for cheap.

Good luck!

Greg

[LooseChange]

Okay, if I disconnect the primary... what would this tell me? And how would I plug it in??

I disconnected the filaments and still a short.
Used the ohm meter and measured all the primaries and secondaries every possible combination. All pairs of wires that are labeled on the schematic read exactly as written. Found no shorts between any connection that should not have one. All leads that don't lead to ground read open on my ohm meter.
Is there any other test I can do or should I start shopping for a PT???
Thanks!

[LooseChange]

By the way... This amp is a pain to work on... Rats nest!!!

[sluckey]

Okay, if I disconnect the primary... what would this tell me?

If it no longer blows the fuse, it means the primary has shorted turns. If it still blows, look at the wiring between the line cord and the PT (and there ain't much there). My money is on a bad PT.

By the way... This amp is a pain to work on... Rats nest!!!

I agree. I owned a brand new one in '66. All a po' boy could afford. Replaced the cheesy OT within a year. I still don't get it why they are so popular these days. They were affordable back in the day even though they weren't quite loud enough for a teen club, but they could cover most VFW  and AmVets clubs. It was a cheesy design with cheap non-standard parts and even cheaper cab. You had to use Sears replacement parts except for resistors and caps. I can't believe that anyone that ever owned one back in the '60s could have a different opinion of this amp. Guys, this ain't boutique stuff. It's was a cheap, affordable comsumer product that captured the market that couldn't afford Fender. But it's a toob amp and I guess eBay has made it a collectors item.

Sorry for the rant, but it's the truth.

PS... Remember the song "Riders on the Storm" by the Doors? This amp could nail the thunder claps in that song, if the reverb still worked. Just sing "Riders on the storm" then reach over and smash the top of the amp head with your fist. Perfect!   

[SoundmasterG]

Okay, if I disconnect the primary... what would this tell me?

If it no longer blows the fuse, it means the primary has shorted turns. If it still blows, look at the wiring between the line cord and the PT (and there ain't much there). My money is on a bad PT.

PS... Remember the song "Riders on the Storm" by the Doors? This amp could nail the thunder claps in that song, if the reverb still worked. Just sing "Riders on the storm" then reach over and smash the top of the amp head with your fist. Perfect!   

LOL< thats about right....the reverb on these is pretty damn cheesy when it does work. These are a rats nest and a pain to work on for sure. I think the attraction is that they do sound pretty good, and the sound is unique. They can be made to sound better, but they're still a cheap POS...haha.

I vote for a bad PT also....

Greg

{EDIT-- quote-nesting -PRR}

[terminalgs]

big image, hope its not too big:

http://www.dougcircuits.com/images/silvertone1484wideguts.jpg

[terminalgs]

BTW,  my last post with the pics is my 1484.  the choke was replaced at some point, and the tech decided to mount it on the top of the chassis near the tubes ..  thats why its missing from the pic.

[sluckey]

Wow! What a nightmare! I had forgotten just how bad it really was.

[LooseChange]

I am ordering the PT and I hope I don't have to dismantle the entire amp to install.  Then I'll find out what else in broken.  Hope nothing!

[DummyLoad]

weber cloned the OT of that amp...

https://taweber.powweb.com/store/magnetic.htm

model WOT1484.

[LooseChange]

Yeah, thanks. I'll probably need that too.

[SoundmasterG]

It can't be an exact clone as he lists it as having 4 ohm and 8 ohm outputs, and as a 50 watt OT....the stock Silvertone 1484 and 1485 have an OT with a 2.6 ohm and a 4 ohm tap, and there is no way that little thing is a 50 watt OT. I've never seen one in person, and I respect Ted and his products a lot, but I have to wonder at that one.

Greg

[PRR]

> It can't be an exact clone...

...because magnetic iron and electric insulation has changed since the 1960s.

You fake it the best you can.

> he lists it as having 4 ohm and 8 ohm outputs

It is possible to do that without any great change in the performance (perhaps less than production variation was on the original).

> there is no way that little thing is a 50 watt OT.

Iron power handling depends on frequency. I have no doubt it can handle 50 Watts at 440Hz, probably 220Hz. If at 82Hz at high power it only makes ugly-waves, well, that's part of "the sound", no?

[LooseChange]

The amp is only wired for the 4 ohm tap. The other tap from the OT is not connected to anything.
The PT should be here on Monday... It looks like it's going to be a pain to install.

[loogie]

I worked on one of those not too long ago and I used Weber replacement OT.  It was a little taller than the original (or whatever was in there), but it worked great.

The only problem I had was with the vertical circuit card.  It was kind of fragile and the terminal strips were lightly riveted in.  It took only a gentle nudge to cause them to separate from the card. 

I think I used the ground switch to select the output impedance.  It seemed pretty stout.  I actually liked the way the amp sounded at the end.  The reverb is unique.  Definitely not Fendery or in any way springy for that matter, but it lent an interesting texture that I liked.  Probably a stretch to call it reverb. 

[loogie]

Here's a closeup of the Masonite board and the terminal strips.

[LooseChange]

That masonite is closer to compressed cardboard... Look at it wrong and it breaks. Thanks Loogie.