Transformerless amp work-around. Will it work?

[dancemyth]

OK, don't shoot me for asking the question.  I know this is a topic that makes you roll your eyes.

I totally get why transformerless amps are dangerous, but hear me out.

I picked up a bunch of All American Five radios, that I want to convert into inexpensive table-top tube-based bluetooth players to give as gifts.  I'm completely gutting them, replacing the speakers, and putting in a simple 2-tube hifi amp.  Essentially I'm just using the vintage shell and the chassis.  I'm trying to save a few bucks, which is why I'd prefer not to buy power transformers if I don't need to, but obviously - especially since they will be gifts - safety is my primary concern.

So here's the question:  Rather than put in an admittedly-inexpensive 120v/120v (I'm in the US) isolation transformer, what if I build the power supply as follows: 3 prong-plug, ground wire connected to chassis, hot wire to on/off switch to fuse to one input of a solid state rectifier; neutral wire to other input of SSR.  The B+ connection of the SSR will be positive regardless of which wire was hot coming from the outlet, and the amp-side ground will be always be ground, so even if the outlet was wired backwards, there's no connection to the chassis, and there's no way the amp-side connections can be inverted.  So isn't this just as safe as an isolation transformer? 

(note I still will be adding a 6.3v filament transformer.  No way I'm putting in capacitor-driven voltage divider or zeners or mosfets to step down 120v to drive the heater.  The 120v is just to give me a B+ of ~170v)

I figure I'm missing something.  So educate me.

Even if this does work from a safety perspective, is there a noise consideration?  I know iso's also provide noise suppression, but I got lost in discussions of faraday shields, and could not tell if this was an issue/benefit for small music amps.

Thanks
Dan

[sluckey]

     

[PRR]

With OUT external audio cord connectors, it has the same risk as when it was an AM radio. If the user does not open or crack the totally insulating case, and does not drop it in the bathtub, it is unlikely to kill.

Putting RCA or phone-plugs on is what makes transformerless sets deadly.

I'm sure 99 of of 100 of your gift recipients will not know a tube from a $2 class D amp.

[DummyLoad]

Rather than put in an admittedly-inexpensive 120v/120v (I'm in the US) isolation transformer

spend the few dollars and don't kill, injure or mainme your friends and family. you know the risk, miswired receptacle, 2 prong ground bypass adapters. any one of the them is trouble, they exist and then chassis is at 120V potential. there's a reason for the safety code we have today and pardon the language, but why in the hell would you look to circumvent it?

this and this.

+ the blue tooth module.

--pete

[d95err]

Also good to know that fuses are there to protect electric equipment from being damaged in case of faults.

However, they don’t do much to stop humans from being damaged...

[JB]

  I'm trying to save a few bucks,


 - safety is my primary concern.

Only one of those statements can be true.  You've answered your own question I think.


 

[shooter]

Only one of those statements can be true

I love logic puzzles, especially on crayon nights 

[dancemyth]

Thanks for the replies.  I'm not arguing, but I really want to understand this.  How is the design I described any more dangerous than a transformer?  If I have 2 power leads connected to a full-wave rectifier, with no chassis connection (other than the power ground), how is that any different from those 2 leads being connected to a transformer winding?  Is it because of possible diode failure?

PRR, first let me say, I have learned an awful lot reading your posts.  To answer your question, my relatives and friends all know I build tube amps (yes, the normal kind, with transformers, etc).  They will appreciate the "coolness" and pseudo-vintageness of having a couple of tubes doing the work in there, even if it's totally impractical.  Your observation would probably also apply to most guitarists, but that doesn't stop us from building tube amps.

However, it can't be dangerous.  And I don't agree that I can't save money without decreasing safety.  Obviously the vast majority of cost decisions we make in building amplifiers (paper in oil caps, anyone?) have no effect on safety.

[shooter]

start here, I didn't do well in English 

https://www.allaboutcircuits.com/technical-articles/transformer-isolation/

[pdf64]

How were you planning on these amps interfacing with the outside world?
eg is the bluetooth stuff all within the amp cabinet, or will there be an audio jack and hardwire connection of some sort?

Generally, electronic components connected to wall voltage usually require special design and approval. Consider lightning strikes to power lines etc, and transformer isolation makes good sense.

[dancemyth]

Thanks guys for your continued patience with me.  I want to understand this.
Shooter, your article illustrates the types of risks that I understand.  But from what I can tell, they stem from the power supply line and the following circuit sharing the same ground (which could be the chassis).  In that situation, if the neutral line is actually hot, then the common ground (and maybe the chassis) is also hot.  But in my design, the power line doesn't share a common ground with the following circuit (and is not connected to the chassis), so it wouldn't matter if the wrong lead was hot.
(Picture attached)


pdf64, I was planning on putting in a 3.5mm audio jack, but, again, the chassis is not being used as a power ground, so I don't see how high voltage could get onto the chassis.  Your point about power line power problems (lightning strikes, etc) is a good one.  I assumed 1000v/3A diodes would be safe, but maybe that is the weak spot in my thinking.

[DummyLoad]

you do know that neutral is bonded to ground at your service panel, right? and if your ground rod is not in place or has poor conduction then there exists a potential difference from earth to neutral. this is a very, very bad idea for something that has connections to humans and will interface with other properly grounded equipment.

build it and see for yourself. you've been given the thumbs down by several here already unanimously. let us know how it works out.

and lastly... think of it this way, if someone thought that this was a good idea, then we'd see a lot of equipment built this way - and no, this idea or design is not original, so now i ask, what do you think about this idea knowing that now?


respectfully,


--pete

[sluckey]

Read this message and then the entire thread. Maybe this may appeal to you...

     http://el34world.com/Forum/index.php?topic=25597.msg277868#msg277868

[dancemyth]

Whew, this is harder than I thought it was going to be.
First of all, thank you everyone.
Second, I'm obviously going to use an isolation transformer.  I'm not a knucklehead.  I realize that if everyone is advising me to do that, then clearly they are right and I was wrong.  That was kind of my assumption from the beginning, because clearly, to Pete's point, this was not an original idea.
All of which brings me to point 3: Please, I just want to understand why.  Not because I'm disagreeing, but because I hate not understanding why I'm doing something.
I'm also not lazy, I did a fair amount of reading before I made my original post.  I thought I understood the safety rationale for the isolation transformer, but I don't see how it still applies when the power ground is not shared with the circuit ground and where neither the hot wire or the neutral wire are connected to the chassis.  pdf64 made an excellent point about components that are designed to work with wall voltate, and I think Pete was trying to educate me with his comment about the potential potential difference between earth and ground, but I'm still not getting it.  Pete, I expect that you are too exasperated with me to explain it in a little more detail.  But maybe someone can give me a more complete explanation or a link that will educate me.  Unfortunately, I think my attempt at expanding my understanding has been misinterpreted as either arrogance or argumentativeness.
Steve, thanks for the design suggestion.  You helped me with another project (an organ amp rebuild) on another forum and, like everyone else, I've learned a ton from you.
Truly I have a ton of respect for the jedi masters on this board.
I'm hoping at the end of this, I can smack my forehead and say "Oh, NOW I get it!"

[sluckey]

Steve, thanks for the design suggestion. You helped me with another project (an organ amp rebuild) on another forum

I don't recall ever helping a dancemyth. What was your user name on the other forum?

[shooter]

a quick stab;

Ground 1 and ground 2 in the pic have nothing in common, in ALL modern stuff, they are common, via chassis, maybe a ground lift circuit.......
in your case there could be 1800W (15amp 120vac house circuit) between Ground 1 and 2.  AND they are, by your image "close enough to touch"

[dancemyth]

Steve, I had to look for it.  Turns out it was actually here under the name dsmith9.  I was the one who posted all of the documentation from a 1954 Wurlitzer model 44 amp. (I never did post pics of the finished product).  I didn't realize I had 2 names here.  Maybe I'll go back to that one, since I clearly haven't made any friends using this one.

OK, so Shooter, if Ground 2 is also connected to the chassis, does that solve the problem?  And if not, why isn't this still a problem in the transformer-based design.  I suspect that there is something in the whole "voltage is relative" thing that I'm not getting.  I've been tripped up by that before in my understanding.

One of the things I love about building amps is that at first you don't understand why you are doing anything.  You're just trying to build something that works and not electrocute yourself along the way.  Then you learn about grounding, because you want your 2nd amp to be quieter than your first.  Gradually you understand more and more about "why" you do what you do and not just "how".  Before this little aside, I was working on improving my understanding of power amps.  But this is an interesting aside for me.  Obviously every amp I've built had a power transformer.  At first I thought it was just for the step up in voltage.  Then I came to understand that there is a safety aspect as well.  I definitely understand why the All American Five design is inherently dangerous.  I'm now trying to understand why a properly implemented full-wave rectifier doesn't effectively provide the same isolation as a transformer.

[pdf64]

...I was planning on putting in a 3.5mm audio jack, but, again, the chassis is not being used as a power ground, so I don't see how high voltage could get onto the chassis...

Unless the whole thing met the spec for being 'double insulated' (and it isn't even single insulated), the sleeve of the jack will need to be robustly held at ground potential.

...Your point about power line power problems (lightning strikes, etc) is a good one.  I assumed 1000v/3A diodes would be safe, but maybe that is the weak spot in my thinking.

Lightning is an extreme, non-routine example, used to get the point across. But there are significant kV spikes on the powerline 24/7, eg maybe a neighbour using an arc type welder etc in their workshop. A transformer acts to insulate, isolate and filter those spikes from the equipment's circuitry. Without a transformer, we must eliminate the possibility of a person coming into electrical contact with the circuitry, eg touch an input jack. As regular LBC fuses, 1kV diodes, even resistors can't be relied on to cope with the onslaught from spikes, voltage may well eventually track over anything that's not a sufficiently resilient insulator.

[shooter]

I believe my brother is missing from the picture, he uses 0 tech, phone still has cord, '70s record player........so while "we" can and are safe playing with electricity, Tom might get knocked on his butt. so that alone stops me from "releasing" anything that doesn't at least make minimum current safety. The "why" has enough "proof" that the answer is; "because smart ppl say so" 

pdf64 talks about the parts, do some math on 120vac@20A, you get a really big diode 


 

[dancemyth]

OK guys, thanks for the patience.
I'll use the transformer.  Just because.
And I'll go back to trying to figure out power amps.  At least there I feel like I'm getting smarter.

[shooter]

I'll use the transformer.  Just because.

full discloser, I wanted too do the same thing you asked about went I was as a newbie here, I moved on with transformers 
When I troubleshot for a living the last thing I cared about when I "fixed It" was why, my brain was already past the problem, in my jeep heading for some NC rock climbing.  When I got back, there was ALWAYS another problem waiting 

[dancemyth]

Thanks.  I said that I was not stupid, not arrogant, and not argumentative - but I didn't say I'm not curious, or determined.  So I will take continue with my research, so that eventually I can explain to someone else why transformers confer an inherent safety benefit.  (Partly because I wonder about all those consumer electronic power supplies that lie all around the house). 

[shooter]

because I wonder about

 
switch mode Power supplies is another completely different can of worms 
I have many - too many- cats, curiosity never kil't a one, trucks however........

[66Strat]

Thanks.  I said that I was not stupid, not arrogant, and not argumentative - but I didn't say I'm not curious, or determined.  So I will take continue with my research, so that eventually I can explain to someone else why transformers confer an inherent safety benefit.  (Partly because I wonder about all those consumer electronic power supplies that lie all around the house).

you do know that neutral is bonded to ground at your service panel, right? and if your ground rod is not in place or has poor conduction then there exists a potential difference from earth to neutral. this is a very, very bad idea for something that has connections to humans and will interface with other properly grounded equipment.

build it and see for yourself. you've been given the thumbs down by several here already unanimously. let us know how it works out.

and lastly... think of it this way, if someone thought that this was a good idea, then we'd see a lot of equipment built this way - and no, this idea or design is not original, so now i ask, what do you think about this idea knowing that now?


respectfully,


--pete

The answer is in the first line of DummyLoad's post. The ground wire that you show in your schematic is connected to the neutral AC wire at the electric panel. There is no connection between the transformer secondary leads and the line voltage leads.

[PRR]

> I was planning on putting in a 3.5mm audio jack

That's where you get in trouble. You can hide the chassis inside wood. You can get signal in through wood. You can get sound out through wood. So a basic Radio can be 'safe'. The user never touches anything connected to the power from the wall.

As soon as you add a phono needle, guitar input, line output, the user can be in direct contact with power, either the line directly or (maybe worse) some jack-up from there due to DC conversion. Since all our usual audio connectors have exposed metal. The onboard phono was least dangerous because users need not touch the wires/clips every day, and the very high impedance of crystal and tubes allows blocking caps too small to be lethal.

[Ritchie200]

Here ya go, $33 all in one.  Pass the price increase on to your relatives!


Jim


https://www.tubesandmore.com/products/transformer-hammond-plate-filament-or-bias-63-v-auxiliary

[DummyLoad]

Here ya go, $33 all in one.  Pass the price increase on to your relatives!

Jim

https://www.tubesandmore.com/products/transformer-hammond-plate-filament-or-bias-63-v-auxiliary

jimbob...  these folks have more betta price for hammond $#!t. don't expect same day shipping and whatnot, they take a day or two to ship.

https://www.hawkusa.com/manufacturers/hammond-mfg/enclosures/262e6

dancedude, they are reputable - i have bought from them 4-5x in the last few months.


--pete