Walker's Confounding Lamp - discuss

When I built my system, I wanted to have a lamp on the panel than would light up when there was current flowing through my heater element. I bought some 120v lamps for this.

Here's how it was connected right after installation in the in-progress panel. Red and black come through contactor. Red line is then switched by SSR and feeds lamp and heater element (heater element not plugged in at this time). Black line goes right from contactor to element. Lamp was connected to the switched 120V hot line and neutral.



After hooking it up, I flipped on the contactor. SSR was not even connected to the PID at this time. To my confusion, the lamp illuminated. Eventually I realized that this made sense. The SSR leaks a little current, and the lamp needs very little current to illuminate, so the leaked current was taking it's only route available and was coming through the lamp to neutral.

So, I think, "Well... when I have a much lower resistance path available for that leaked current to go travel through, then it should work." So, I plugged in the heater element. The SSR is still not connected to the PID, but it now looks like this:



When I turn on the contactor, the lamp still lights up.

THAT'S QUESTION #1: why is the little bit of current that leaks through the SSR going through the lamp instead of going through the much lower resistance heater element?

Screw it. I am using the system. I used it like this for a year. The lamp comes on as soon as I activated the contactor, and stays on, fully illuminated, no matter what. PID might be driving SSR, might not, but the lamp is on. Period. The only way to shut the lamp off is to turn off the contactor.

I've been adding things to my panel and tweaking stuff recently. I decided to fix this thing. I got rid of the neutral and connected the lamp fully in parallel to the heater element. With a resistor added to match the lamp so that 120V drops across the lamp and 120V drops across the resistor. This DOES behave in a sane way, but just wait for more insanity.



In this configuration above, the lamp is partly lit when the contactor is on and the heater element is not plugged in and the SSR is not being triggered. This makes sense because of the leakage current from the SSR. When the SSR is triggered, the lamp lights up fully. This all makes total sense.

So, plug the heater element in now:


Everything SEEMS to be functioning perfectly. With the heater plugged in, leakage current from the SSR goes through the element and not the lamp, and the lamp stays dark. When the SSR is triggered and the heater starts to fire, then the lamp comes on fully. Perfect and exactly what I wanted.

Happy with the fact that it seemed to be working, I shut everything off and turned out the lights in the garage.

Wait... What's that!? The lamp is on just BARELY. Hard to see with overhead lights on, but obvious when the garage is dark.

So... this is the REAL confusing part to me.

Why is the lamp on at all when I have this:



The black hot line to the lamp is alive because it's coming from a point prior to the contactor, but the red hot line is DEAD. The contactor is off. There can't be any leakage through the SSR, because there is no source attached to the SSR.

I response to Q #1, the element is providing a path for the black line to "backfeed" your lamp. If you go black through contactor, element, lamp to neutral you've got a complete circuit.

As far as why the lamp was glowing when you turned the lights out, that's a good one. If you weren't using a GFCI I would suspect a VERY high Z path to ground. How long was it between when you opened the contactor and when you turned the lights out?

Also, why not just move the black line to the lamp to the load side of the contactor. That way the power's totally cut.

#1: Electricity flows down ALL paths, not just the one of least resistance.

#2: I'm assuming your lamp is neon? They require very VERY little current. Under 2mA usually. Even with that 50K Ohm resistor, it's still seeing over 2mA. As for why it's glowing after everything's turned off, I'd guess the PID's internal power supply has a cap that's leaking current back over the SSR signal lines.

Incandescent indicator lamps are much less headache for projects like this...

Terminator, at the point where the machines start to take over. Just wait until your Stainless starts to form different humanoid shapes.


I am expecting similar results when I finally fire up my new panel and this will be interesting to watch.

Q1. Yes, I know that electricity flows on all paths, but out of the total current flowing, a prroportional amount will flow on each available path, inversely proportional to the resistance of that path.

In other words... current is being allowed to flow through the lamp AND the element in the original circuit. the vast majority of the current should flow through the element and not the lamp, because there is much much less resistance through the element.

To ask a related question, why does it behave PROPERLY now, but not in the original circuit? From the leaking current's point of view, there is little difference. There is a low resistance path through the elmement and a high resistance path through the lamp.


I do not want the lamp connected to the contactor's switched line because the lamp is on the swinging door of my control panel and the contactor is down in the belly. The swinging door has it's own complete set of wiring that is all tapped from the black incoming feed, and I power everything on the panel door with 120v. I'd like as few connections going from the belly to the door because of the long wires that are necessary when I open the door.

That's why the lamp was originally hooked to 120v. One hot line coming up and out of the belly to the lamp, but then a short connection to the neutral bus mounted on the door.

To jkarp: yes, the lamp is neon and does require very little current to light up, which is why the leakage is capable of making that happen when the SSR is off.

but, the slight illumination when the contactor is off at this time is not coming from the PID. It happens even when the SSR control lines are not connected to the PID!

regarding slight illumination:

perhaps there is some slight capacitive behavior in the SSR, allowing a little charge to be built up in the SSR as the one hot line that is connected swings back and forth across the 0v line? That would result in a bit of a difference in potential between the swinging line and the point connected to the SSR, and would give a super tiny voltage drop across the lamp....

hmmmmmm

Walker said:

regarding slight illumination:

perhaps there is some slight capacitive behavior in the SSR, allowing a little charge to be built up in the SSR as the one hot line that is connected swings back and forth across the 0v line? That would result in a bit of a difference in potential between the swinging line and the point connected to the SSR, and would give a super tiny voltage drop across the lamp....

hmmmmmm

Click to expand...

Most SSRs do have a snubber circuit on the output that contains a cap. Kinda surprised it would be enough to light neon though...

I'm unclear on the initial change from the separate neutral for the lamp. Because the SSR leakage current is fixed, adding the element effectively drains the leakage current and drops the circuit voltage below the neon's threshold.

jkarp said:

Most SSRs do have a snubber circuit on the output that contains a cap. Kinda surprised it would be enough to light neon though...

Click to expand...

Well, this thing is barely lit. Like I mentioned, I never even noticed it until the lights were out and it was completely dark in the garage.

A cap is the only thing I can think of that would cause it to behave like this.

jkarp said:

I'm unclear on the initial change from the separate neutral for the lamp. Because the SSR leakage current is fixed, adding the element effectively drains the leakage current and drops the circuit voltage below the neon's threshold.

Click to expand...

By "unclear", do you mean that you also don't understand why it behaved so oddly at first?

I have puzzled and pondered on this a lot. It was strange enough that I thought others would have fun puzzling and forming their own theories.

The leakage is because of the snubber internal to the SSR. You can purchase snubberless SSR's. Read the following:

http://www.galco.com/techdoc/crdm/d2450_faq.doc

Walker's Confounding Lamp

I would consult the Magic Item Compendium or the Magic User's handbook

... Don't make me get out my 20-sided die, ollllllo ...

My illuminated switches are wired similar to your original drawing. The switches have 120v LEDs.
If I don't have a load (pump or element) plugged it, the LEDs will illuminate. With the load plugged it, the LED's are off when the SSR is off.

Ed

Yup. That's why mine are Confounding Lamps. You got Regular Lamps.

Ohio-Ed said:

My illuminated switches are wired similar to your original drawing. The switches have 120v LEDs.
If I don't have a load (pump or element) plugged it, the LEDs will illuminate. With the load plugged it, the LED's are off when the SSR is off.

Ed

Click to expand...

Exactly. The element acts as a bleeder resistor on the leaking current. This pulls the output voltage below what's needed to drive the LED.

So why didn't Walker see this expected behavior in his original design? There has to be more to that separate neutral he ran as things were "fixed" when he tied to to the same neutral as the element.

I wired my indicator lights to the coil on the contactor. When the contactor is closed, my indicator light is on.

But that doesn't tell you if the element is heating. It just tells you that the contactor is on.

I use an illuminated switch to enable the contactor, but I wanted a lamp to tell me when the was heat was actually being applied.

@jkarp: note that the "fix" was to connect across 240V and not use a neutral. (No neutral to my elements).

This might be a neat application for a CT coil. You could just use it to turn the light on or set up current/wattage metering...that would be cool to know how many Watts you used in a brewday.

Walker said:

@jkarp: note that the "fix" was to connect across 240V and not use a neutral. (No neutral to my elements).

Click to expand...

OH! There was no mention of 240V in the OP! I was assuming everything was 120V. Well that's obvious then. The element was acting as your 120V hot, lighting the lamp all the time.

Walker said:

But that doesn't tell you if the element is heating. It just tells you that the contactor is on.

I use an illuminated switch to enable the contactor, but I wanted a lamp to tell me when the was heat was actually being applied.

@jkarp: note that the "fix" was to connect across 240V and not use a neutral. (No neutral to my elements).

Click to expand...

That is how mine works and I really like it.
I can tell from a distance if the temp is at or near set point based on how the lamps react.... Solid ON means there is a way to go yet, blinking means getting close or at set point, On/Off with a pause means we're good to go.

Ed

jkarp said:

OH! There was no mention of 240V in the OP! I was assuming everything was 120V. Well that's obvious then. The element was acting as your 120V hot, lighting the lamp all the time.

Click to expand...

I thought the diagrams made the 120v vs 240v pretty clear.

And, yes, it is obvious now that I look at my own pictures. Here's the original arrangement, simplified to remove the SSR since it was off:



And now that I have read about the RC network in the SSR, it makes sense that the lamp is getting a tiny trickle of current even when the contactor is off in the latest wiring.

All mysteries solved. I feel stupid for not recognizing the original problem, but at least I can laugh at my own mistakes and stupidity.

Walker said:

I thought the diagrams made the 120v vs 240v pretty clear.

And, yes, it is obvious now that I look at my own pictures. Here's the original arrangement, simplified to remove the SSR since it was off:

And now that I have read about the RC network in the SSR, it makes sense that the lamp is getting a tiny trickle of current even when the contactor is off in the latest wiring.

All mysteries solved. I feel stupid for not recognizing the original problem, but at least I can laugh at my own mistakes and stupidity.

Click to expand...

I told you weeks ago the answer would be simple just not simple enough for us to see :cross:

"Not simple enough to see" or "too simpke to see" ?

Walker said:

I thought the diagrams made the 120v vs 240v pretty clear.

Click to expand...

I'm sure Picasso's art made PERFECT sense to him too.