Relay Problems

[micky01]

I've made a circut using realPCB which is all good and what not, however, their is a problem with my relays. I'll try my best to describe how they are set up.

The relays are SPST Solid 240VAC 3A - More info - ( http://www.altronics.com.au/index.asp?area=item&id=S4410 )

I have basicly emulated the circut from the picaxe manual.

V+ is connected directly to +pin on the relay.

The -Pin on the relay is connected to the collector on the transistor (Can remember off the top of my head, but it was a BCX somthing) and also a protection diode is across this track, connecting to V+.

Connecting to the Base on the transistor is the output pin on a picaxe with a 10k resistor.
And on the Emitter is the 0V line.
The problem is, as soon as power is applied to the board, the relays are on consistantly on, that is without the picaxe in or if power is on the board but not to the picaxe.
Can anyone shed some light where i may have gone wrong?

Thanks

 

[moxhamj]

The inputs could be very high impedence, and when the transistor is off, one of the inputs is at 5V but the one connected to the transistor is floating, and in reality is connected by a 1000M resistors (dirt on the case etc) to anything. To ensure it is off, maybe pull that floating input high with a 100k to 5V.

Do you need the transtor at all? I have driven these directly with picaxes. But I have also found they do leak slightly when off (flouro lights glow a bit) and I found an old fashioned relay worked better.

 

[Cruiser]

I agree with Dr_Acula, you should be able to drive this relay direct off the PICAXE. It will look just like an LED. You won't even need a dropping resistor. The specs claim 2000ohms "coil resistance" which means only 2.5mA at 5 volts. Jaycar have an almost identical device SY4080.

I do the same with their much bigger 40A solid state relay.

Too easy!

Chris.

 

[MiB1986]

Hello,

Can you post your schematic, or pcb artwork, i drive relays all the time using a 08M,

The setup is as follows, output pin from picaxe through a 10k resistor into the base of a TIP31A, collector linked through the relay with a protection diode to ground, and emitter connected to 12v

Hope this helps,

 

[Rickharris]

A simple quick fault finding "trick" with transistors is to short the base and emitter with a screwdriver etc.

This will turn the transistor off. If the relay drops out you know the problem is with the drive to the transistor - if not then the problem is either the transistor circuit or the relay connections.

 

[gengis]

"The problem is, as soon as power is applied to the board, the relays are on consistantly on, that is without the picaxe in or if power is on the board but not to the picaxe."

There's two choices: the driving transistor is leaking or biased on slightly. A resistor between base and ground would help that - 10 K should do it.

Breaking the connection to the "coil" should also turn it off if that's the case.

SSR's have "snubber" networks built into them to protect the internal SCR's from transients and prevent noise. If the relay has no load on the output, it may check bad because the snubbers pass a small amount of current around the SCR's (thyristors)all the time

Checking them with a high impedance meter may indicate they are turned on when they are not. A small load like a neon pilot lamp may indicate it is on when it is not. They need a load of some minimum value to indicate correctly.

 

[micky01]

Thanks for all your posts guys, very helpfull and sorry I didn't reply sooner (site was down over here).
It turns out that Dr_Acula was right. I don’t need a transistor at all, or a 10k resistor.
But should I leave the protection diode their? Or is it not necessary since it’s a solid state relay and interference will be minimal?

 

[moxhamj]

Good to hear it is working. Site has been down here too. The diode is used in relay drive circuits to prevent the back EMF from the coil zapping things, but since a solid state relay doesn't have a coil the diode shouldn't be needed.

 

[boriz]

... unless you are driving an inductive load?

 

[premelec]

Not unless it's arcing across the isolation barrier inside the SSR! Usually you are just driving an LED in series with a resistor in a SSRly....

 

[Michael 2727]

<b>Important </b> SSRs or Solid State Relays have an LED on the input, if you blow up the LED you get a <b>PAPERWEIGHT </b>

The input may say 3V to 40V DC <b>BUT IT MUST BE CURRENT LIMITED </b>
Many SSRs state 16mA as a MAX current, most of them will work from as little as 2 to 3 mA. I usually settle for whatever voltage and 10mA current. Happy medium.

 

[KMoffett]

Michael 2727,

Unless I'm mistaken, all SSR's that &quot;state&quot; a range of input voltages (i.e. 3 to 40vdc) already have an internal current limiting resistor in series with the LED. In your example, if it states &quot;3-40vdc input&quot; and a &quot;maximum current of 16mA&quot;, the internal resistor is 2.4K...(40.0v-1.7Vled)/16ma. If there is no internal resistor, the spec should just give the the maximum current and the Vfwd of the LED. The maximum input voltage would be pretty much irrelevant, as I could put 1KV on a 100K (10W) in series with the the LED and only pass 10mA through the LED. If I'm wrong, could you point to SSR's that you're look at.

 

[Michael 2727]

Most SSRs do have an internal resistor <b>but </b> it is usually a small value under 100R or so, e.g. 30 to 50 Ohms.

I was once breadboarding a new $47.00 SSR, as I was inserting the +POS 12V wire to a Switch/Resistor on the input side when the wire slipped and briefly hit the +Pos input on the way past, <b> Instant Paperweight </b> .
I was in a hurry to test the thing and only switching 24VAC, so silly me had everything ON.
YOU NEED A RESISTOR ALWAYS.

 

[Dippy]

Before this ends in an oh-yes-it-is / oh-no-it-isn't argument I think we'd better look at components - and mke sure you're talking about the same type of device.

A number of the 'pcb' mount SSRs have quite large value Rs in them. eg:
<A href='http://www.farnell.com/datasheets/54578.pdf' Target=_Blank>External Web Link</a>.
I think you'll admit that Crydom is a well known manuf?
Crouzet too for 'surface' style.

I guess the rule-of-thumb giveaway is the 'coil' side voltage range figure.

Bottom line, as usual (stifled yawn), read the blooming Data Sheet.
I'm sure even those cheap'n'nasty semi-anonymous Beijing bangers have a Data Sheet somewhere.

 

[Tom2000]

And while we're on the subject of SSRs, CP Clare makes a very nice little dual SPST NO one in an 8 pin DIP package. Look around for the PAA110 or LAA110.

It's just the thing to replace reed relays in MCU applications, saving a passel of board real estate in the process.

And for these, you do need current limiting resistors between the Picaxe outputs and the relay LED inputs.

Tom

 

[KMoffett]

I was building a little control circult with an 8M. I had used every I/O pin, when I realized that I needed a second ADC input. Rather than redesign and reprogram for a larger Picaxe, I wound up adding a little SPDT 8-pin DIP Mosfet relay, Panasonic AQW614EH, to switch (by an external control) the one 8M ADC input available, between the two analog sources.

<A href='http://pewa.panasonic.com/pcsd/product/pmos/pdf_cat/aqw61_eh.pdf' Target=_Blank>External Web Link</a>
http://pewa.panasonic.com/pcsd/product/pmos/pdf_cat/aqw61_eh.pdf

Pins 7 and 5 were tied together and to the one ADC input. Pins 6 and 8 to the two analog sources (two pot wipers). Pin 4 to ground, pins 2 and 3 tied together, and pin 1 (yes, through a 3.3K series resistor) to a externally controlled 5v line.