Go Down

Topic: Switch mode power supply and solid state relay (Read 2 times) previous topic - next topic

bld

I am pondering on using a switch mode power supply with a solid state relay, bit is that a good idea?

My concern is the solid state not disconnecting it completely, and maybe allows the power supply to be powered up a bit.

I have tested this with a smaller switch mode power supply (500mA wallwart), and with a LED on, the LED is still lit after turning the solid state relay off.

But will something like this scale up? My plan was to use it with a 30A switch mode power supply for my 3D printer, but when turned off, it should be turned completely off!

The reason for the solid state relay is that it uses so little power to turn on, and the mechanical using more power than I got avalible on the printer without the external power turned on.
captain-slow.dk | non contagious!

pwillard

solid state relays are prediminatly AC devices..  Not DC.   That explains why it won't shut off.  A zero transition is needed.

bld


solid state relays are prediminatly AC devices..  Not DC.   That explains why it won't shut off.  A zero transition is needed.

So they can't be used with anything that turns AC into DC?

And when used on AC with an AC device on, they still leaks a bit.
captain-slow.dk | non contagious!

retrolefty

Quote
My concern is the solid state not disconnecting it completely, and maybe allows the power supply to be powered up a bit.


Your concern is proper. Solid State relays normally assume there is a purely resistive load being switched on and off, where voltage and current are always in phase. When wired to an inductive load there will be some phase angle difference between voltage and current. Not sure what kind of reactance load a switching power supply might offer the SSR and it might be variable depending on the load current the power supply is providing. This kind of switching problem can often be solved by wiring a 'snubber circuit' across the SSR output terminals and is comprised of a series capacitor and resistor. The values might have to be calculated for your specific load device. There are some SSR I've seen that have an internal snupper circuit installed. I'm sure if you search the web you should be able to find some useful information.


http://electricalandelectronics.org/2009/04/17/design-of-snubber-circuits-for-thyristor-protection/

bld


Your concern is proper. Solid State relays normally assume there is a purely resistive load being switched on and off, where voltage and current are always in phase. When wired to an inductive load there will be some phase angle difference between voltage and current. Not sure what kind of reactance load a switching power supply might offer the SSR and it might be variable depending on the load current the power supply is providing. This kind of switching problem can often be solved by wiring a 'snubber circuit' across the SSR output terminals and is comprised of a series capacitor and resistor. The values might have to be calculated for your specific load device. There are some SSR I've seen that have an internal snupper circuit installed. I'm sure if you search the web you should be able to find some useful information.


http://electricalandelectronics.org/2009/04/17/design-of-snubber-circuits-for-thyristor-protection/

But to calculate it according to load might be a problem, because load is somewhere between 400mA and 28A, depending on what is going on. And even "stand-by" load can change +/-200mA

Maybe it would be better to find a way to power a relay instead. Maybe a small charger circuit with a battery to help pull the relay, and turn the external power supply on.
captain-slow.dk | non contagious!

dc42

Is it a standard ATX computer power supply? If so, you can turn the supply on/off through its own control input. Of course, a small part of the supply stays active when you turn it off.
Formal verification of safety-critical software, software development, and electronic design and prototyping. See http://www.eschertech.com. Please do not ask for unpaid help via PM, use the forum.

bld


Is it a standard ATX computer power supply? If so, you can turn the supply on/off through its own control input. Of course, a small part of the supply stays active when you turn it off.
No, it isn't. The problem is not 'something' inside the power supply being turned on, the problem is the 12V that comes out of it. I want it to be turned off when the power supply is off, so it won't feed the stepper drivers anything, and not be able to accidently power any of the heaters.
captain-slow.dk | non contagious!

dc42

I'm a fan of SSRs for AC switching, but you have a legitimate concern here. Without trying it, it's hard to know what your SMPS will do when you connect an SSR in the off-state in series with it.

In the past I've used a triac to switch AC, where the triac is driven by pulses at a frequency of a few tens of kHz, through a pulse transformer to provide isolation. This may be a workable solution for you.
Formal verification of safety-critical software, software development, and electronic design and prototyping. See http://www.eschertech.com. Please do not ask for unpaid help via PM, use the forum.

Grumpy_Mike

Yes a SSR will have a small trickle current when off. The way to get round this powering your power supply and producing a voltage is to put a resistive load across the input of the power supply to short out this current. Something like an incandescent light bulb is a useful load, which also gives you an indication of when your power supply is on.

bld


Yes a SSR will have a small trickle current when off. The way to get round this powering your power supply and producing a voltage is to put a resistive load across the input of the power supply to short out this current. Something like an incandescent light bulb is a useful load, which also gives you an indication of when your power supply is on.

Will something like a little 0.6W be enough, or is it better to take a more powerful one?
captain-slow.dk | non contagious!

dc42



Yes a SSR will have a small trickle current when off. The way to get round this powering your power supply and producing a voltage is to put a resistive load across the input of the power supply to short out this current. Something like an incandescent light bulb is a useful load, which also gives you an indication of when your power supply is on.

Will something like a little 0.6W be enough, or is it better to take a more powerful one?


From the data sheet of your SSR, find its maximum off-state current @ 230VAC. Then pick a bulb that takes more current than that (say double) at its rated wattage. For example, if the off-state current is 10mA maximum, you need a bulb of around 5W or more.
Formal verification of safety-critical software, software development, and electronic design and prototyping. See http://www.eschertech.com. Please do not ask for unpaid help via PM, use the forum.

Grumpy_Mike

Quote
From the data sheet of your SSR, find its maximum off-state current @ 230VAC. Then pick a bulb that takes more current than that (say double) at its rated wattage. For example, if the off-state current is 10mA maximum, you need a bulb of around 5W or more.


Close but no coconut.

There are two things to consider, one is the leakage current of the SSR and the other thing is the minimum holding voltage of your switch mode power supply. The load you need to use has to be such that when the leakage current flows down it the voltage developed across it is less than your power supplie's minimum holding voltage.

It is likely that you don't know your power supplie's minimum holding voltage, but it could be as low as 30V. Therefore it is impossible to tell you a load value you will most likely have to experiment.

dc42


Quote
From the data sheet of your SSR, find its maximum off-state current @ 230VAC. Then pick a bulb that takes more current than that (say double) at its rated wattage. For example, if the off-state current is 10mA maximum, you need a bulb of around 5W or more.


Close but no coconut.

There are two things to consider, one is the leakage current of the SSR and the other thing is the minimum holding voltage of your switch mode power supply. The load you need to use has to be such that when the leakage current flows down it the voltage developed across it is less than your power supplie's minimum holding voltage.

It is likely that you don't know your power supplie's minimum holding voltage, but it could be as low as 30V. Therefore it is impossible to tell you a load value you will most likely have to experiment.


The third thing to consider (that works in your favour) is that the resistance of an incandescent bulb when cold is less than 1/10 of its resistance when hot. Using my suggestion to select a bulb that normally takes double the off-state current, the self-heating of the bulb when it is passing the off-state current is going to be very low indeed. My guess is that its resistance will be not much higher than one-tenth of the hot resistance, which means that the input voltage to the power supply will only be about 12v. That should be low enough.
Formal verification of safety-critical software, software development, and electronic design and prototyping. See http://www.eschertech.com. Please do not ask for unpaid help via PM, use the forum.

Grumpy_Mike

Quote
The third thing to consider (that works in your favour) is that the resistance of an incandescent bulb when cold is less than 1/10 of its resistance when hot.


Was that not blindingly obvious, sorry. But I was only talking about the DC load not about an incandescent lamp rating. I had assumed, maybe naively, that if you wanted to use a bulb you would measure the resistance with a meter.
Just to underline this you measure the resistance with a meter when the bulb is off.

dc42


Quote
The third thing to consider (that works in your favour) is that the resistance of an incandescent bulb when cold is less than 1/10 of its resistance when hot.


Was that not blindingly obvious, sorry. But I was only talking about the DC load not about an incandescent lamp rating. I had assumed, maybe naively, that if you wanted to use a bulb you would measure the resistance with a meter.
Just to underline this you measure the resistance with a meter when the bulb is off.


Blindingly obvious to you perhaps, maybe not to the OP or other readers of this thread (if I was replying only to you, I would send a PM instead of post a reply). I didn't wish to assume that the OP had a selection of bulbs that he could measure the resistance of, that is why I suggested a way of choosing a bulb wattage that was likely to work.

An alternative would be to use a PTC thermistor in parallel with the power supply input, for example this one http://uk.rs-online.com/web/p/thermistor/1911888/ should be OK up to an off-state current of 50mA.
Formal verification of safety-critical software, software development, and electronic design and prototyping. See http://www.eschertech.com. Please do not ask for unpaid help via PM, use the forum.

Go Up