SCR and relay: like an ordinary diode... doesn´t work

[host666]

scr fires relay

Hello to all reading this post and this forum, my best regards.

I´m trying to put a SCR to work like a SCR but he denies to work like one.

What I want to do: activate an 48V DC coil relay using the AC mains and a control SCR.
When the SCR is triggered, the circuit becomes "crazy" and it appears to start auto-oscillating. The relay turn on-off forever with a period of 1s aprox. Note that using an ordinary 1n4007 diode the relay works ok, with it´s nominal coil voltage level. The control circuit is just a DC level to fire the SCR, nothing unusual, don´t use optocouplers, neither zero-crossing detectors, just a simple and effective DC control level...

Please take a look at the atached circuit, then you will understand my problem and how I could fix it.

Help! Can´t understand whatos happening! Could it be that the capacitor voltage level became "bigger" than the AK voltage and the SCR stop conducting? Could it be the latch current of the SCR? (tried a lot of them, none worked).

Why I want to use the mains to power the relay and why I don´t use an 120vac coil relay?
1- Miniature Power PCB relays with a 120VAC coil is very dificult to find.
2- I have space limitations at the circuit
3- I have current limitations at the circuit power supply (transformerlless), can´t drive a relay without wasting a lot of energy
4- the circuit have to drawn very low power when i´ts in stand-by mode.

Thank you to everyone, can´t wait to make this circuit working...[/b]

 

[max0412]

half working relay diode

One thing I see is that you are only drawing 7.8mA. The datasheet says the minimum holding current at 40 deg C is 10mA. Try placing a resistor cathode to ground to provide at least 10 mA. Place a UF diode across your relay inductive transient is generated when a relay is energized .A large rate of rise voltage (dv/dt) can trigger an SCR into conduction.I also don't see how you are resetting the SCR once it is latched. A simple RC filter to limit gate current and suppress any transients on the gate couldn't hurt either.

 

[johnyaya]

how to use scr as a relay

From the ON Semi data sheet:
Reverse Peak Gate Voltage (VGRM) = 5.0 V

The input control appears that it is referenced to the neg side of C1. When C1 charges to more than 5V the gate voltage then becomes negative with respect to the SCR cathode. Because C1 can potentially charge to 170v I'm surprised that the SCR is not burnt. For the above circuit to work the gate control voltage needs to be referenced to the SCR cathode (or the pos side of C1).

 

[host666]

what is ordinary diode

Hey johnyaya, max0412 thanks for the prompt reply.

About the latch and holding current, I´ve already put a resistor drawing about 50ma, do not work either. The relay keeps, as it seems, auto-oscilating (after the SCR fire, with the SCR off, the relay stays off).

About the transient generated from the relay, it´s about 5 times the relay coil, or 240v. The SCR is rated at 400V.

This is the protoype circuit. It should turn on-off a load upon a valid IR receive.

I think that the half-rectifier transformerless power supply give me the requirements for driving the SCR. The scr gate current, relative to the catode, is positive.

When the PIC receives a valid IR comun., it will fire the small SCR to drive the DC coil relay. It´s a pure +5VDC to the gate, no phase or zero crossing detection. Simple as it meant to be.

Just a note: when the capacitor is took out off the circuit, it seems to work fine. The relay chater ´cause the pulsating DC after the SCR, but the waveform at the scope is OK. When a large capacitor is put again, the circuit becomes crazy again...

I use DC coil relay cause mini 120VAC PCB relays are dificult to get, and I have limitations on size and budget. Using a TRIAC to drive the load is not an option, because of its higher price compared to the relay and because it will be very hot, needing something to irradiate the heat (more space used than a relay).

Sometime ago I used a MPSA44 (400v) transistor to do the job, worked for some time, but one of them exploded, think thats because of the power at the transistor when it´s conducting (or I did something wrong). Simulating with the computer tells me that all is fine regarding the transistor power and current, but it exploded.

Now this circuit is the cheapest..... and useless...

Thanks,

Host666

 

[host666]

how to make scr work as diode

Hey johnyaya, max0412 thanks for the help.

I was asking this question at the electronics zone forum, and Pebe from Scotland helped me to fix the problem.

That was the problem, (from Pebe):

"Posted: Sun Aug 19, 2007 11:30 am Post subject:

--------------------------------------------------------------------------------

Your problem is that when SCR fires, C6 charges to peak mains voltage and so the gate is reversed biased and turns off. When C6 discharges almost to zero SCR fires again. That gives the effect you have.

If you rewire the SCR with its cathode grounded, that will solve the problem. You need the diode to isolate the SCR during negative half cycles of the mains.

http://www.electronic-circuits-diagrams.com/forum/viewtopic.php?p=5947#5947"

The solution is attached as an image.

Thank you for your patient and your prompt replies.

Host666

 

[RF-OM]

ordinary diode

First of all you need to provide DC path to the gate circuit. Did you check it? I think that after you tune this circuit everything should be okay.

Best regards