Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Need help isolating one part of a circuit from another

Status
Not open for further replies.

Projectman

New Member
Hi. I have a display made out of el-wire. For those who don't know about el-wire, it is electroluminescent wire that can be bent into shapes. It is powered by high frequency AC power and draws very little current. For example, you might use a 9 volt battery with an inverter to provide 100 v AC power at 2000 hz. It draws something like .002 amps per meter.

My circuit has several inverters connected to various strands of el-wire, and uses relays to turn on and off the inverters. I attached a simplified diagram.

Because the inverters/relays are located a few yards from the el-wire display, a run the AC power to the el-wire through a multi-conductor cable. I have only enough conductors to run one wire to each strand of el-wire, so I connected the second wires together and ran them back to the inverters in a common wire. So far, so good - it works great.

My problem is that occasionally, one of the el-wire strands may short out. This is probably because I'm not great at soldering and the leads are tiny. It could also happen from a faulty inverter or some other reason. When this happens, it is normally not a big deal - one strand of el-wire goes out - such is life.

However, part of the circuit has strands of el-wire going on and off in sequence. Inverter one goes on, then one goes off while two goes on simultaneously, and so on. When one of these went out, it took the next one with it.

The inverters don't go off instantly when the power is cut by the relay - it looks like they hold their charge for a fraction of a second. So, my uneducated theory is that the short somehow goes through the inverters during that fraction of a second, shorting out the next strand.

Is there any way to prevent this? If it was a DC circuit, I would put a bunch of diodes in before connecting the commons together to isolate them. How is this done with AC power?

Thanks for any suggestions.
 

Attachments

  • El-wire circuit.pdf
    19.6 KB · Views: 164
Hi,

I guess the problem is the inverter. Interrupting the 100V line the inverter continues working producing overvoltage with no load connected.

You might interrupt the primary side. If it takes too long time for the inverter to start interrupting the 100V output will suffice if you keep the interter working with a load connected.

Here is a suggestion how to switch the inverter output with a load resistor connected via the NC terminals of the relay. The current fow will be 1.8mA, taking care of no overvoltage.

Boncuk
 

Attachments

  • 100V-SCH.gif
    100V-SCH.gif
    10.5 KB · Views: 181
  • 100V-BRD.gif
    100V-BRD.gif
    20.3 KB · Views: 162
Thank you very much for replying, Boncuk, but I don't understand the suggestion. The load (el-wire) is always connected to the inverter. The relay powers the inverter when open and cuts power when closed. Inverters don't like to operate with no load, so I cut the DC power to the inverter to turn off the el-wire, rather than leaving the inverter running and interrupting the AC power from the inverter.

Are you suggesting somehow discharging the inverter more quickly by putting a dummy load on the normally closed side? I don't understand how that would work. Also, I can't follow the diagram you attached as I'm not sure what most of the symbols mean - is there any way to label it? My understanding of this is fairly basic, so unfortunately I'll need more detail to understand the response.

Thanks again for the help.
 
Hi projectman,

my suggested circuit leaves the inverter running, no matter if you connect or disconnect the load.

When no load is connected and the relay is in the NC-position R2 (56KΩ) is the load the inverter has to drive. That resistor takes care of sufficient current flow if no "real" load is connected.

Switching the relay to the NO position the connected load is driven by the inverter.

Disabling an unregulated inverter might cause high voltage spikes when energized. R2 takes care of spikes not being stronger than the rated voltage and current of the inverter.

More or less you leave the inverter operating at all times and just switch different loads, one being a "dummy" load (R2) and the other one the real load (led stripes).

Boncuk

P.S. I forgot to draw one connection at the relay. Terminals 11 and 21 have to be connected for higher current flow. They are located in a straight line on the PCB.
 
Last edited:
Thanks again, Boncuk. I've thought about that approach and didn't pursue it because I didn't know how to make the dummy load. You may be right that flipping the inverters on and off could create a power spike and damage the el-wire. I would really like to understand how to build your circuit, but I can't interpret the diagram. I can tell what the battery, resisters and diode are, but I really can't tell what is what. Perhaps "K" is the relay, but which side is NO or NC? I have no idea what Q1, J2, J3, D2, or S1 are. I'm used to diagrams using symbols like this: https://www.arrl.org/tis/info/pdf/8402019.pdf. Do you have a link to an article that would decode the symbols used in your diagram?

That said, need a different solution for the near term. I have already built my circuit, and I need to use it at an event coming up in a few weeks. I have it working pretty well, but I just want to see if there is a way to prevent a short in one strand of el-wire (whether caused by a spike from the inveter or some other cause) from burning out another. I'm pretty sure this has happened because of the order that the el-wire strands burned out.

For example, if inverters 1-5 go on and off in sequence, and the el-wire connected to inverter 3 burns out for example, there is a good chance that after a couple of more passes of the sequence, the el-wire connected to inverter 4 will burn out. I think due to the physical nature of the inverters, the fact that the relay 3 turns off at the exact instant that relay 4 turns on, and the fact that all of the el-wire strands share one common wire (that is, one side of the inverters are connected together) there may be a brief time when el-wire 3 is connected to el-wire 4.

So, for example, is there a simple circuit I could put between the relay and the inverter that would delay power going to the inverter for a fraction of a second after the relay closes? Or, is there something I could put between the inverter and the e-wire to prevent AC power from flowing from one inverter to another for the fraction of a second they are both powered?

Thanks again for any help.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top