Relay controlling Relay, Won't dropout 1

[palindrome]

Hello, My question deals with a GE wireless Outdoor Motion-Sensor Light Adapter. The part number is PIR612A. Now as all good engineers, I using this product differently than its intended design. Instead of using it to sense motion and turn on a 120 VAC light I would like to use it to sense motion and energize a second relay which in turn operates a 12VDC 2A motor. The Sensor Adapter has the following spec's labeled on its side: 125VAC 1.2A 60Hz 150W TUNGSTEN. The wireless unit is powered by a 9volt battery. It appears that the 9Volt battery actuates the relay in the sensor adapter.

Problem: The sensor adapter has no hesitation in energizing the second relay but as some of you have already guessed the second relay doesn't drop out when the sensor adapter relay de-energizes. Through some experimenting it looks like the sensor adapter relay doesn't like ANY DC load. I tried a 60mA DC light with the same results (so I'm guessing a snubber or supression won't be of help). The sensor adapter energizes but won't de-energize. The 12 volts DC that I want to switch on and off is still present. I don't know what type of relay is in the sensor adapter because it is sealed up - I'm guessing electromechanical because it has and audible click. Maybe a REED Relay? Anyone here work for GE Lighting Division that could shed some light - pardon the pun! Any insight, thoughts or experience would be greatly appreciated. Last resort would be to break into the adapter and reverse engineer it. It always seems to be the simplest of things......or just things that appear to be simple that hangs us up!!!! Thanks

 

[jbartos]

Suggestion: Visit

http://www.safetycentral.com/homandbusals.html

etc. for an alternate product

 

[busbar]

Could the sensor have a time-delay on “120V” contact opening for the lamp? {Adjustable, maybe?}

 

[palindrome]

There is a selectable time delay, either 1 minute or 10 minutes, that works fine when 120VAC is switched through the the sensor adapter relay. It's when the sensor adapter relay tries to switch a 12VDC load that the sensor adapter relay never drops out. If I remove the load after the slected delay expires, the relay will drop out and energizes as expected when motion is sensed again. But then it won't drop out this next time either unless the load is once again removed.

To verify the device is functioning properly I hooked it up, as designed, to a 120 VAC lamp and it all functions as expected. It makes me believe that the type relay or how it's wired has something to do with it. GE customer service can't provide any information on internal part spec's or operation other than how the instructions describe.

Maybe this will provide some additional insight: If I connect a digital ohm meter across the contacts, the meter switches between indicating an "open" and "closed" contact - repeatedly without a problem when sensing and waiting for the delay to time out. I'm thinking 50mA to 200mA shouldn't be a problem???! Thanks

 

[ve7brz]

I'm guessing the load controlling device is a traic, possibly actuated in turn by a reed relay. If the device works properly with an AC load but stays latched on with a DC load you definitely have a triac.

 

[Pavel2]

Hi Palindrome
I am probably missing the point entirely, but why not use a relay with an 120V ac coil? Not being an engineer myself, maybe that's too easy solution?
Good luck
Pavel

 

[ve7brz]

Hi, after thinking about this for ten minutes I've come to the conclusion that you definitely have a triac controlling the load circuit. The triac gate is powered by the load current and is itself controlled by a reed relay. Look at a lamp dimmer circuit and substitute switch (or relay) contacts for the variable resistor.

The only way around this is, as Pavel2 just suggested, is to use an AC powered relay.

 

[mysterymike]

I agree, its a triac.

 

[ve7brz]

On second thought, if you must use a DC supply for your relay there are other ways around your problem. Unfortunately the explanations would be lengthly so I'll wait for more information before proceeding.

 

[palindrome]

First, thanks to all for the alternative product websites and suggestions as well as identifying what the device might be.

Unfortunately without knowing what the output controlling device is, if there is one at all, experimentation would be necessary if specifications or documentation, on the available alternative products, was scarce.

One of the reasons I liked this particular product was its advertised "wireless" feature. Somewhat of a misnomer because it WAS designed for switching 120VAC! But being powered from a 9 volt DC battery it looked like it had potential for my application. My application is sensing motion at a remote location and activatng a 12V 2A DC motor. The intended power source is a 12VDC car battery (solar charged) with a 9 volt regulator for the sensors power.

TRIAC: So we believe it is a TRIAC because a TRIAC conducts current during the both alternations of the AC cycle and that would be more efficient for lighting as opposed to an SCR and one half cycle operation. And a thyristor would make sense, especially a photo-SCR/TRIAC since this is a motion sensor. The TRIAC probably operates a REED relay since the 'click' is heard.

I think I understand why, but not entirely sure, that when an ohm meter was used to verify 'ON' - 'OFF' operation the ohm meter indicated and open and closed relay. (When I did go back and check myself on this 'closed' was actually 450 ohms.) Is this because the ohm-meter current was less than the holding current of the TRIAC so it responded to the time delay function ? - BUT that would mean that the TRIAC/REED RELAY COIL/RELAY CONTACTs were somehow ALL coupled together?

ve7brz, Since I need DC operation the only way to reset the TRIAC is to reduce the current to less than the 'holding current' (opening load) or drop the supply voltage (also reduces the holding curent). I don't know if the time delay function on my sensor is used with a gate controlled switch (GCS) - I'm guessing no? So how does a TRIAC turn off by way of the time delay function if it must remove the supply voltage to reset - could it be a gate controlled switch after all?

Obviously I haven't done much design with Thyristors but I recall, from text book days, something about "backdooring" an SCR to turn it off also - is this a form of supplying current at the output of the device to reduce the holding current? Thanks again to all, you have provide GREAT input!

 

[ve7brz]

You're on the right track. The circuitry inside your sensor will be a 400+ ohm resistor in series with the reed relay contacts. These components control the current flow between one side of your 'load' line and the triac gate. Keep in mind that the reed relay controls the triac, not visa-versa.

You have four ways to overcome the 'self-latching' property of the triac and you've already mentioned three of them.

The simplest is to keep the current flow so low that you're below the holding threshold of the triac and simply sensing the operation of the reed relay. This is what you saw with your ohmmeter. That small current could be used to drive a transistor or sensed with an Op-Amp or FET to drive your DC relay.

The second method is to use what you referred to as 'back-dooring' the triac (although I've never heard that phrase before). The concept here is to charge a capacitor and then dump it into the circuit such that the current flow through the triac is temporarily reduced or halted. The configuration in this case is to put a (large) non-polarized capacitor in the circuit with appropriate P and N channel FETs connecting both terminals to both sides of the supply. You then alternate the FET drives so that the capacitor is effectively 'short-circuiting' the supply several times a second.

Obviously you need a diode and capacitor at your relay to keep it energized during these 'short-circuits', and you also need some impedance back to your power source, either a resistor or an inductor or both.

The third method is very similar to the above, but uses a 'chopper' to momentarily interupt the current flow. The complexity created by this method relates to keeping the dv/dt below the triac's threshold when restoring the power.

Then, lastly, the obvious; use an inverter. Note that the inverter must produce a good quality sine wave or you'll exceed your dv/dt.

 

[palindrome]

An inverter would be nice but I think I can keep cost down and battery capacity up with some of the other methods! Thanks again, I've got a much clearer picture and probably some new/old questions where I failed my education - lol.

 

[jbartos]

Suggestion: There may be the triac or the switch formed by Q1 and Q2 transistors forming a switch of regenerative type and offers a high impedance until the control voltage drop reaches a preset level. As the level is reached, the switch is triggered and conducts. The trigger point may be determined by the motion sensor. When the motion sensor is on and the regenerative switch cannot trigger. As a result, the relay across it remains actuated, releasing only when the regenerative switch conducts, shorting out the coil voltage and permitting the contacts that supply the external load to open. The regenerative switch voltage is half-wave AC rectified by a diode, and the switch may thus be triggered on each positive half-cycle. This device function may be linked to garage door openers.