Signal relays

Is this a one-off or something that has to work in a variety of circumstances? Do you have numbers for what the switch portion must withstand? If it's like the low-voltage, low-power of most thermostat situations there are a number of possibilities, including optically isolated triac ICs that might work. Does this 3V limitation come from your device (seems rather low)?

nces?

like

er

Yes this is a one off. I'm building a custom thermostat to replace my wall thermostat. I know these furnace control circuits are 24VAC, and the curr ent to the furnace (heat and fan) is probably low, but I'm not sure about t he A/C contactor, which is the main reason for my hesitation with using a s ignal relay. I assume the relay in the thermostat directly controls the po wer to the A/C contactor, but that's what I'm not sure of.

I did look at SSRs, but I'm leery about those because when they fail, they often fail closed.

The 3V limitation is actually a preference because the microcontroller used is 3.3V. I could have another regulator to provide relay coil power, but thought it might be nice to have one regulator.

Well, a 5V signal relay would open and close at 3V; the closing/activation time is slower, tho. But contacts for most SIGNAL relays are not rated or designed for POWER applications that you seem to allude to.

Well - if it's a one-off, then you can measure it, right? And you can design the circuitry to be bulletproof. A well protected solid state device (in this kind of application) should be far more reliable than a relay. If you're not so comfortable with hardware design, then do yourself a favor and buy one that does what you need - they're not that expensive.

sign

this

not

that

I could measure it, but also wanted to know the design convention since I w ant it to be compatible with future systems, and with others if anyone else wants one. I also like things to reliable, so knowing how the rest of the system is designed, and what HVAC manufactures typically do, is helpful.

Honestly I'm getting scared away from the signal relays. I think a 2A GP r elay would be more than sufficient, based on a 1.0A rating I've seen other commercial thermostats.

I'm quite comfortable with H/W design, and I'm designing my own because the only ones on the market I like are $500, but I also find it fun. This one has an Ethernet jack and ARM processor and can have all the off-the-wall f eatures I want because it's my design.

Incidentally I already have a unit running on the bench on a custom PCB. I t's working fine but due to some minor issues, I will be having a new set o f PCBs fabricated, and the relays are part of the problem. The ones I have are maybe a little too big and the coil power dissipation is affecting the ambient temp. sensor.

On a sunny day (Tue, 27 Aug 2013 19:30:58 +0000 (UTC)) it happened Frank Miles wrote in :

Right, look up 'solid state relay'.

I'm aware of SSRs, and for AC I know I have two options: TRIAC and MOSFET o utput. The MOSFET output versions seem better since this voltage range sho uld have less power dissipation. I'm just leery about protecting the outpu t from damage. Take this for example:

What should go on the output to make it immune to damage from inductive loa ds or ESD? Just a TVS, or RC snubber? This is one area I'm not so knowled geable.

I'm all for no moving parts but sometimes you can't beat the reliability an d robustness of metal contacts.

Those get hotter and are less reliable than a relay.

A neat trick to reduce the power consumption in a relay is to use a capacitor parallel with a resistor in series with the relay. When the relay is powered the capacitor delivers the current to make the relay swith over. After that only a small current is required to hold the relay. The minium hold current is specified for most relays.

On a sunny day (Wed, 28 Aug 2013 12:33:27 GMT) it happened snipped-for-privacy@puntnl.niks (Nico Coesel) wrote in :

Well they were used in PLC gas control systems, and met all reliability specs. You need to select the right current type. It is true the voltage drop is a bit more, unless you have seen relay contacts after several thousand switches. Relays (the normal ones) are unreliable as h*ll.

Even more components. The solid state relays are easy to drive. and you can make your own too of course.

On a sunny day (Wed, 28 Aug 2013 03:35:48 -0700 (PDT)) it happened snipped-for-privacy@yahoo.com wrote in :

You need to say what your current is that you want to switch, frequency too.

I once had to switch an inductive load with a small one (fluorescent tube), and used VDR across the device.

SCRs will conduct on huge spikes, snubbers can be used, I like VDRs for simplicity. For a resistive load I would not worry to much, the normal spikes on the mains would be more than from switching your load I'd guess, those can be kV.... Snubbers consume a bit of power when the relay or triac is off...

Relays are very very unreliable, really, both the low power ones, and the high power ones, limited lifetime.

It's the 24VAC control for a residential HVAC. Seems like 1.0A is the max based on some commercial thermostats having that as the max. rating.

I believe the worst load in my case is the contactor in the outdoor A/C unit. Good luck finding coil specs on those. I do need to just measure mine and add some safety factor, but lately the A/C has been very busy.

I'm leaning towards 60V 2.5A,

BUT..a mechanical relay IS solid state. Mostly metal if i remember correctly..

Just because it's metal doesn't mean it's solid. For example, mechanical mercury relays.

"The term is often used to contrast with the earlier technologies of vacuum and gas-discharge tube devices and it is also conventional to exclude elec tro-mechanical devices (relays, switches, hard drives and other devices wit h moving parts) from the term solid state."

Nonsense, they're off-the-shelf.

Those have not been made for at least 20 years..

Yea, yea, yea; i know, teach.

I've handled a few of the things before. Awesome to pick up a piece of otherwise normal, if oddly shaped, electromechanical equipment, and HAVE IT SLOSH AROUND WHEN YOU SHAKE IT. :)

Tim

We used tons (literally) of the things in systems for plastics machinery.

Best regards, Spehro Pefhany