Run indicator / power-light off of rectifier tube 5vac ?

[kevtronic]

Is there any reason why the indicator/power-light can't or shouldn't be run off of the 5vac rectifier tube filament juice?
I searched a bit on the web and couldn't really find any info on it, so I figured I'd see if I could spark some interest here on a kind of off topic I guess...
Just kind of curious what others may think/know about it or perhaps someone has even done it or doing it now type thing

[Enzo]

The 5vAC that heats a rectifier tube is also sitting at B+ voltage, so your pilot lamp would have 500vDC on its terminals. it wouldn't hurt the bulb, but if anyone were to touch the socket, he'd get a 500v shock - potentially lethal.

If the amp lacks a tube rectifier, but has the 5v winding unused, then sure, why not?

Assuming the tube rectifier implies it is a tube amp, meaning 6v heater power is available, then the question is: why would we need to make it more complicated? Just power the lamp off the heater string like they have done for decades.

[kevtronic]

A high voltage DC shock is a very good reason not to do it...
I didn't even think about that to be honest.
Kind of solves that curious wonder moment pretty quick - lol...

[R.G.]

I've just been doing the opposite - figuring out how to run low voltage indicator bulbs off the AC power line without transformers.

The simple thing to do is to swap out the indicator bulb for a neon bulb that naturally runs off the AC mains voltage. There are lots of possibilities out there.

Another thing is what I did - develop a capacitor-limited voltage dropper to run those low voltage incandescents off the line. In my case I was trying to power 28V incandescents that were already in a vintage amp. Took some work, but with the main ingredients of an X2 safety cap, a bridge rectifier and a zener, I have them running directly from the AC power line. The hardest part of the solution so far has been trying to keep safety-standard creepage and clearance distances.

[Chuck H]

Considering that resistance circuits (that tend to fail open) could be in series prior to any chassis mount sockets I don't see an issue. With the possible exception that limiting with capacitors (that tend to short rather than go open) is your concern. A fuse seems like overkill for what should be a very simple and inexpensive circuit in an "appliance". Is limiting with a capacitor really less costly than dropping with a resistor? Or is this more of a novel exercise?

[olddawg]

What's wrong with using an LED pilot lamp off the AC mains. They have them cheap as plastic assemblies that drop into standard size chassis holes. If you want to make your own a diode (or bridge rectifier) and a current limiting resistor (and maybe a 100uf cap) will work fine. You just have to do the math. If the 5v secondary is unused you can do the same thing there. People run a neon or LED lamp off the mains after the switch all the time in amps that use a 6.3 vac lamp to try to free up enough heater current from an existing PT when adding another tube for a gain stage, reverb, etc.

[R.G.]

Considering that resistance circuits (that tend to fail open) could be in series prior to any chassis mount sockets I don't see an issue. With the possible exception that limiting with capacitors (that tend to short rather than go open) is your concern. A fuse seems like overkill for what should be a very simple and inexpensive circuit in an "appliance". Is limiting with a capacitor really less costly than dropping with a resistor? Or is this more of a novel exercise?

Resistors convert voltage drop to heat. Capacitors are essentially non-dissipating. The particular problem I was solving had several odd characteristics that conspired to make a search for something different worthwhile.

I was replacing the silly switch assembly for Thomas Organ Vox amps. Thomas switched the speaker line for "standby", two indicator lights of 28V/1W nominal bulbs in a non-relampable assembly, and the AC to the power transform4er all in one assembly, running the indicator lights off DC from the power supply. They also grounded the amp in such a way that any ripple on the DC fed to the lights appeared as current ripple and a slight hum in the ground to the power amp chassis. All that taken together eventually led me to think about running the lights on AC, not the amp's DC. After all incandescent bulbs are inherently AC as well.

Resistors work, but running a dropping resistor from 120vac to 24vac dissipated about 4W of heat and had mounting issues in the enclosure. I figured I could do it better. First iteration was to replace the lights with neon assemblies, as identical appearing replacements are available. Vox owners did the obvious and complained that neon wasn't "original". Sigh.

So I went haring off looking for less-dissipative ways to use the original assemblies. I thought about phase control ( too much electrical noise ), off-line LED drivers ( too complicated and tricky to get running right ), power factor correctors ( not really applicable), tiny transformers (not tiny enough, and expensive) and finally capacitive dropper power supplies.

Switching power supplies effectively require caps attached to the AC mains for interference suppression. Safety rated X and Y caps were developed for this purpose. They're metalized film caps which have been tested and proven to be self healing. Spikes may puncture the insulation, but this burns off the metalization around the puncture and does not form a persistent short. They're what line reverse "death caps" should have been, but weren't.

I found suitable X2 safety caps for under $1. A wirewound resistor is cheaper, but not a whole lot, and it eats several watts while sitting there heating up your AC power switch and death cap. And it fit well with the other strange requirements for this assembly.

[Dave H]

Is limiting with a capacitor really less costly than dropping with a resistor? Or is this more of a novel exercise?

It's not the cost. The bulb is 6.3V 0.15A. Running it from 120V mains would dissipate 17W in the resistor.

Edit: Simulpost

[J M Fahey]

running a dropping resistor from 120vac to 24vac dissipated about 4W of heat and had mounting issues in the enclosure.

Base problem is that filament lamps eat too much current, period.

I live/work in a decrepit >100 y.o. 19 room house, which is electrically divided in 6 sections , which as prevention I keep unpowered unless I am doing something *there* , I already had an electrical fire which burnt 1/3 of the house, so ....
Problem is that by day I go to some particular room and turn on a drill, winder, saw, whatever ... and only after loooonnngggg seconds my sloooowwww brain realizes itīs not a switch/cable/mains socket/plug failure but simply I forgot to flip that area switch ON.

So in many outlets, at least one per room, and in a visible place, I drilled the mains socket plate and installed a LED ... since Neons are all by extinct ... at least in standard shops.
Oh , they have all kind of Neon indicated *thingies* (99.99% Made in China of course) , but no sane human being is supposed to buy spare neon bulbs now, is it?
So counter monkeys become nervous and smile a lot when I ask them for spare Neon bulbs, and talk loud (and winking, thinking I donīt see them) at each other asking "you donīt think we have spare neon bulbs, do you?" ..... "oh Sir, sorry, we just run out of them, I guess weīll get them together with our next delivery of unicorn dust and vampire dried blood"
Ok, they donīt actually say that, but itīs the feeling.

So I drill a 5mm hole and mount a high efficiency clear Led (forget standard diffused ones) and grind the front lens flat, so I almost reach the actual Led crystal.

I feed them 220VAC through a 220k 1/4W resistor (it dissipates 0.22W so itīs warm but fine) and a 1N4007 diode.
Led gets puny 1mA with 50% duty cycle, and emits a pinhead sized very bright dot of light which is very visible in daylight, way more than a regular Neon bulb.
Presence/absence of Mains voltage is unmistakable.

On 120V mains you can use a 120k resistor (or splurge and use 100k ) for same average current and it will dissipate mere 120/140 mW , piece of cake.

As simple as can be, works every time.

Surprisingly bright and contrasty, so small that at normal room distances you donīt see it like a round Led, but a shapeless bright point, like a Star.
Itīs smaller than eye resolution so we canīt make out its shape.

[mozz]

If it's a old amp, i put a 6.3v LED bulb in the socket, Mouser sells them. If i can save 150ma draw on the power trans, it can't hurt. Something i am building from scratch, pilot lamp goes on the 120v line.

[loudthud]

Did Thomas ever make any of those Vox amps for export? All the ones I've seen had a single primary on the PT which makes me think North America only. Did you try the capacitive dropper at 50Hz?

[R.G.]

I'm not sure, but I believe Thomas Organ Vox was US only. It makes sense - Thomas was a US company, and bought the US distribution rights from JMI IIRC. You're right - I don't see any indication of a higher voltage or lower frequency on any of the TV literature.

The difference with a capacitive dropper on 50Hz is that you have to use a 1.2x bigger cap to get the same current. It's the cap's impedance that is limiting current from the line and thus dropping the voltage. For a bigger voltage, say 240 instead of 120, you use a half-size cap. So for 240, 50Hz, the cap is 0.6 times the 120/60 value.

You need to put a small resistor in series with the cap - I used 160 ohms - and a PTC fuse just in case the cap doesn't live up to its safety rating and does short. Well, I had to, anyway. I'm paranoid about line faults. I wound up using a full wave bridge, a 22F/50V cap, and a 27V zener in addition to the cap and resistors, bringing the bulb back to DC operation. I did this to have the zener catch any start up transients. Simulation showed that the bulb died on the first half cycle for some values of the timing of the main AC switch turning on. If you happen to start the AC at just the wrong moment, the voltage spike is dumped into the 28V bulb and burns out the filament. So I put in some clamping for the startup transient. Seems to work.

I simulated this thing all I could, trying various faults for this or that part shorted or open. Any fault for a part other than the cap produced a mildly larger line current, perhaps 50ma instead of the 40ma design value. No surprise there. With the 160 ohm resistors, a shorted cap led to about 2A of line current, 27W in the resistors. I used 1W resistors, so something would get HOT. A PTC fuse rated for maybe 250ma stopped this before they get too hot. That was probably paranoia on my part again. This circuit sits right after a 2A line fuse for the whole amp, which ought to catch a cap fault, and the cap is designed not to fail shorted, but still, a $0.40 PTC is pretty cheap insurance.

Capacitive dropper power supplies are pretty common in consumer goods that run off the AC line and have double insulation ratings. You can run a microcontroller and some switches and lights as well as a triac AC power switch or two from them. Very handy.