High voltage Light Bulb Matrix Display (4 Digits)

Im not sure if "Matrix Display Digit" is the right phrase but let me explain what I am hoping to do.

I have a vintage scoreboard with 4 digits. Each digits contains 27 120v bulbs. To do this with relays would require over 100 relays. I want to keep the incandescent screw in bulbs to keep the aesthetic of the score board (No LEDs).

What I am wondering is if there is any way anybody can think of controlling 27 120v Light Bulbs without relays. The digits are more or less small matrix displays.

If that is not at all possible then could I do it if I replaced them with 12v bulbs?

What wattage are the 120v bulbs your using?

TRIACs are the typical way. For a simple on/off you don't have to worry about phase cutting and all that - just one TRIAC with optocoupler and some resistors per light bulb, offering a pretty cheap solution.

Add a bunch of shift registers in the mix and you're pretty much done. One small PCB for every 8 lights (they're cheap to make), each with 8 TRIACs, 8 optocouplers and 1 shift register. Those PCBs in turn can be daisy chained, 14 of them and you can control each lamp individually. At low wattage bulbs you don't even need heat sinks but you will need to offer sufficient ventilation as together they will produce quite some heat.

You probably can even matrix them, though not sure if matrixing works well for incandescent bulbs.

Are you sure you need a relay for each bulb? I ask because you say they are "more or less" a matrix, but each contains 27 bulbs. 27 is a strange number for a matrix. The only shape that fits would be 3 X 9 which is very tall & narrow. Unless it is not a complete matrix and there are spaces with no bulb? Can you post a picture?

I think you might not need a relay/triac for every bulb. If you can form the digits 0 to 9 from groups of bulbs, you only need a relay for each group. Your display would operate more like a 7-segment display, with each segment consisting of a group of bulbs.

Do they look like this, by any chance?

*****
*   *
*   *
*   *
*****
*   *
*   *
*   *
*****

That's 27 bulbs and clearly represents a 7-seg display.

it is a 4x7 grid of bulbs with 2 blank spots in the grid (27 was a typo, should be 26). I can't group the bulbs because that patterns art like classic 7 segment displays. Each bulb must be controlled separately.

I can put whatever wattage bulbs I need to in the sockets.

I am completely new to Triac thyristors. I've read about them a bit and I think I understand the basic premise.

My questions are:

What she difference between a MOSFET and a Triac?

Why would you need resisters if I am using incandescent bulbs?

What are shift registers and why would I need them? Is it instead of using so many IO pins on the Arduino?

Do shift registers use I2C or something of that sort that allows them to be daisy chained?

Could you send me a link to a triac that would fit the bill?

Resistors are for things like the optocouplers and TRIAC gate current limiting.

TRIACs and MOSFETs are very different components. Both used for switching, the first for AC, the second for DC, and a TRIAC will remain on until the current drops to zero (which happens 100 or 120 times a second for regular mains AC), typically used for phase control but for you it's simpler. Just on/off.

Shift registers indeed as pin expansion. You need lots of outputs. They're using a kind of SPI.

Lots of TRIACs out there. The BT136-600 is a common one and will do great. There may be even cheaper ones that work for you as you don't need to switch that much current.

So for $0.30 I can switch 600v at 4 amps? With a 3.3v signal? Almost seems too good to be true haha

When you say "with optocoupler" what do you mean?

How do I figure out what sort of resisters I would need?

Where would they be wired in if they are being used for the triacs?

Circuit example is in the link above and thousands more in Google. Your questions start to scare me: line voltage is no joke and you have to know what you're doing. I have the feeling you don't.

Those triacs are indeed not expensive, I paid like $2-3 for each but maybe not your kind of dollars. Could have gotten it for $0.3 but didn't want domestic versions (mainland copy) to switch 3A. The more expensive ones are Western imports.

I do understand line voltage and definitely I will not be running 600v through these, I was only surprised that such as small component could handle that much power. The first line was a bit of a joke.
I have tons of experience with home electrical systems. I know exactly how to deal with the big stuff. I am just new though to lower voltage electrical components.

I do want to thank you for the help! You have saved me TONS of time and money!

My last question (hopefully) is:

Is the optocoupler to stop the triac from "turning on" on when the Arduino isn't sending 5v? To stop noise from activating it?

Opto-couplers, or Opto-isolators are for safety. They physically isolate the low voltage dc circuit from the high voltage ac circuit. They remove the need for a common ground between the DC and AC circuits. Inside the opto-isolator is an ir emitter led and a phototransistor, separated from each other by a non-conducting medium which prevents any voltage spikes from entering the low voltage circuit. When a small DC current flows through the ir emitter, the phototransistor is switched on by the light and allows current to flow through it.

PaulRB:
Opto-couplers, or Opto-isolators are for safety. They physically isolate the low voltage dc circuit from the high voltage ac circuit. They remove the need for a common ground between the DC and AC circuits. Inside the opto-isolator is an ir emitter led and a phototransistor, separated from each other by a non-conducting medium which prevents any voltage spikes from entering the low voltage circuit. When a small DC current flows through the ir emitter, the phototransistor is switched on by the light and allows current to flow through it.

It's been a while, but I have a couple questions

From what I understand, when you use an optocoupler to to activate the gate of a triac, you use the 120v source to do so. From what it seams, your activating a switch with a switch. Why wouldn't I just use an optocoupler and take out the triac completely (I don't plan on doing so, Im just trying to understand why).

Theoretically, If I wanted, could I use a completely separate DC power supply to power the gate?

Is a SSR just a simple Triac/Optocoupler circuit inside a housing?

Lastly what are some options for switching a DC load. I would prefer not to use relays.

There are different kinds of opto-isolator. In most, the ir led activates a transistor. A transistor can only allow current to pass in one direction (DC). This type is no good for controlling a triac. But in some opto-isolators, the ir activates a small triac. This small triac can allow a very small ac current to flow and control a larger triac which can handle much more power. This type is probably what you have seen on Google search.

You can't activate a triac with a DC voltage, no. Assuming you mean the gate of a triac. Strictly speaking, the triac might be activated for the positive part of the AC waveform, giving you half power at most. Also you should understand that if you try to switch an AC voltage with a transistor, it might damaged by the negative part of the AC waveform.

Yes, a SSR is essentially just a opto-isolator (triac type) controlling a high power triac.

For large dc currents, you can use a MOSFET.

"You can't activate a triac with a DC voltage, no. Assuming you mean the gate of a triac."

Why can't you activate a triac with DC. Is it because the gate would be a completely isolated circuit from MT1 and 2? or just because you could not want to use a TRIAC to switch DC because it would never turn off?

"Strictly speaking, the triac might be activated for the positive part of the AC waveform, giving you half power at most."

This makes it sound like you won't get full power from activating with AC? is that correct?

We are getting close to the limits of my knowledge of the theory of triacs!

What I'm trying to say is that a DC voltage would only switch on the triac in one direction, I think. The triac would still be able to switch off. Its when you use a triac to switch a DC current that it won't switch off.

Very good question!

They're always talking about the four quadrants, where the gate voltage may be positive/negative with respect to the two anodes, and also the anodes with respect to each other. Some switch in all quadrants, others not. So it seems a DC voltage can switch at least certain TRIACs to conduct in either direction.

However using the voltage that it switches for switching (like in the common AC phase cutting application) is I think normally the most convenient, and easiest to keep low and high voltage separated.

All Triacs will switch with a DC control signal. Not all are "four quadrant" meaning that they will switch on both polarities of the AC with a DC trigger of either polarity, some only switch on three of the quadrants. The missing quadrant is usually negative on MT2, positive on gate.

An opto-triac such as the MOC3021 is specified as a trigger device, not intended to pass continuous current; once it triggers, the power Triac to which it is connected also triggers so that there is no longer any voltage across it.

You cannot multiplex incandescent bulbs for two reasons. One is that they conduct in either direction, so that power applied to any particular row and column will see myriad alternate paths through other three-lamp paths; the alternate paths may not light, but will overload the drive circuits.

If you put a diode in series with each lamp, you can multiplex them but will require substantially more than the nominal voltage to achieve adequate brightness and you have to ensure that the multiplexer never stops on any particular phase or it will promptly burn out lamps on that row.

Ahh, Ok that makes sense

Thanks so much for all the help!
I greatly appreciate it