I need 4 solid state relays to turn on/off 4 x 100watt 120vac bulbs used in a incubator for heating (slow not pwm). I would like something that can be mounted and doesn't look like a toy any suggestions
Crydom Solid State Relay ?
I've used these Sharps on several heater type projects, worked well so far.
There are many SSRs, either new or e-bay. The snubber circuit, isolation, finger-safe cover are all included. Your 3.3 amp application can have up to 33 momentary inrush, since you are using incandescent light bulbs. A 25 amp SSR can usually handle inrush like that. See:
Solid state relay
as an example.
Since 100 watt incandescent light bulbs are quickly being phased out in the USA, have you considered using a small resistive heater instead? They have no inrush and are readily available.
Paulcs:
There are many SSRs, either new or e-bay. The snubber circuit, isolation, finger-safe cover are all included. Your 3.3 amp application can have up to 33 momentary inrush, since you are using incandescent light bulbs. A 25 amp SSR can usually handle inrush like that. See:
Solid state relay
as an example.Since 100 watt incandescent light bulbs are quickly being phased out in the USA, have you considered using a small resistive heater instead? They have no inrush and are readily available.
im planning on using 4 relays (one for each bulb). I have considered switching to ceramic bulbs but im also considering using a ldr to check the bulbs for failures. I suppose I could install a CT's and just monitor the amps instead to test for a failed ceramic bulb. (I actually like that idea).
At some point I hope to have a 25w-40w-60w-75w set up as one bulb will probably be on 90% of the time. The other bulbs will kick in after the door has been opened and a quick burst of heat is required. (no idea of real watt numbers until I do a full temp test and for that I need the relays)
I didn't like the idea of using 120v with the cheap arduino relay boards (they are rated to 240vac but they look like toys). so I will get 4 of the style you suggested.
My only other question is weather they can run directly from a arduino pin or do they require a a little more power to operate.
I am currently implementing a "smart security light" that will control a 110v 100W incandescent "living room light". I am using an S108T02 ($4.95 at my local shop, also available from Sparkfun). I am controlling it directly with an Arduino Uno digital out signal. It is working well. I am using a 330 ohm current limiting resistor in the digital output circuit - in series with the ST108T02. I measured the current draw at about 10 mA, well below the 20 mA max per pin Arduino Uno value. After considering several electro-mechanical relays I was very pleased to find this single solid state component that does the whole job - no additional power sources, transistors, TRIACs, etc.
gpop1:
they are rated to 240vac but they look like toys
I take it you haven't taken apart many devices that use electromechanical relays to switch mains power, have you?
Believe me, those relays can handle the voltage just fine - provided the current rating is kept in mind; though for your application, the inrush current would likely overload them.
Size isn't everything.
outsider:
I've used these Sharps on several heater type projects, worked well so far.http://www.sharp-world.com/products/device/lineup/data/pdf/datasheet/s102s02_e.pdf
I second the idea to look into the Sharps. I've pretty much stopped using electromechanical relays and Crydom style SSR's since discovering them for High Voltage projects.
im planning on using 4 relays (one for each bulb)....
...At some point I hope to have a 25w-40w-60w-75w set up...
That shouldn't be necessary. Most heaters, furnaces, air conditioners, and refrigerators are controlled by simply switching on & off.
I've heard of 2-stage systems but that's not automatically a more efficient or more effective design. Virtually no one uses PWM or any kind of continuously-variable design. The heat doesn't change instantly anyway, so you can't really give it "just the right amount of heat" with PWM.
It's also standard to include some hysteresis or "swing". For example, the heat might come on at some set temperature, and then not shut off 'till it's one degree over that temperature (or however tight you want to make it). Then, it doesn't come on again 'till the temperature falls one degree back to the set temperature. That keeps it from "chattering" and switching on & off several times per second, etc.
One advantage of zero crossing SSRs like the Sharps is that they generate very little electrical noise. When electromechanical relays open at somewhere other than the low voltage part of the sine wave they generate tremendous RF noise. Many times mysterious random control system "hiccups" and failures have been attributed to this. If you have a system running 24-7 that is the last thing you want.
On the other hand you might consider is what happens if an SS relay sticks ON. This is one of the common failure modes of SSRs. Often designers will have one electromechanical relay source power to all the control SSRs. This way, if your incubator gets too hot, and turning off the SSRs doesn't slow down the heating, this Safety relay turns off the heat.
cdj15:
I am currently implementing a "smart security light" that will control a 110v 100W incandescent "living room light". I am using an S108T02 ($4.95 at my local shop, also available from Sparkfun). I am controlling it directly with an Arduino Uno digital out signal. It is working well. I am using a 330 ohm current limiting resistor in the digital output circuit - in series with the ST108T02. I measured the current draw at about 10 mA, well below the 20 mA max per pin Arduino Uno value. After considering several electro-mechanical relays I was very pleased to find this single solid state component that does the whole job - no additional power sources, transistors, TRIACs, etc.
thanks for the details that really helps
DVDdoug:
That shouldn't be necessary. Most heaters, furnaces, air conditioners, and refrigerators are controlled by simply switching on & off.I've heard of 2-stage systems but that's not automatically a more efficient or more effective design. Virtually no one uses PWM or any kind of continuously-variable design. The heat doesn't change instantly anyway, so you can't really give it "just the right amount of heat" with PWM.
It's also standard to include some hysteresis or "swing". For example, the heat might come on at some set temperature, and then not shut off 'till it's one degree over that temperature (or however tight you want to make it). Then, it doesn't come on again 'till the temperature falls one degree back to the set temperature. That keeps it from "chattering" and switching on & off several times per second, etc.
im not planning on pwm on this one. I just don't want the constant cycling of a light so the closer the wattage to heat loss the less the lamp will cycle. The only reason I want ssr is they are silent so theres no annoying clicking.
My last build I used peltiers with pwm and that worked well but this time I needed a lot more heat so bulbs seemed a better option. I can always program in hysteria or delays to stop any flickering
Boley:
One advantage of zero crossing SSRs like the Sharps is that they generate very little electrical noise. When electromechanical relays open at somewhere other than the low voltage part of the sine wave they generate tremendous RF noise. Many times mysterious random control system "hiccups" and failures have been attributed to this. If you have a system running 24-7 that is the last thing you want.On the other hand you might consider is what happens if an SS relay sticks ON. This is one of the common failure modes of SSRs. Often designers will have one electromechanical relay source power to all the control SSRs. This way, if your incubator gets too hot, and turning off the SSRs doesn't slow down the heating, this Safety relay turns off the heat.
I have a safety over heat temp sensor built into the design so I think I will follow your advice and add a standard relay. I am already using mosfets for the fans so one extra for a relay wont be a problem
Voltage zero crossing is only sensible for resistive loads though, inductive loads don't benefit
at all. If it were easy to determine the relative phase of current and voltage for a load (even when
its off!) then zero-current switching would be ideal. In fact the best point to switch a mainly inductive
load is at maximum voltage, since this halves the maximum current in the first few mains
cycles under load(*), and reduces the inductive spike at switch off (as maximum voltage is close to
minimum current).
(*) Consider a sine wave voltage starting at t=0 - the first two quadrants are the same voltage sense
leading to rising current in the inductor for two quadrants. In then decays back to near zero by
the end of the full cycle - ie there is a big DC component to the current for the first few cycles.
In practice the inductor may start to saturate too, leading to much larger currents flowing (destructive
levels even), since the start up current reaches twice the normal maximum value.
Now reconsider with a cosine voltage wave starting at t=0. The first quadrant leads to current
build-up, but the second cancels this back to zero and the third causes it to build up in the other
direction - you have started the inductor off directly into the steady-state ac condition (if that
isn't a contradiction in terms).
The sine is voltage zero-crossing, the cosine is max-voltage switching.
See this article for some more insight: http://www2.eefocus.com/设计我最赞/blog/08-10/158191_45d6b.html
MarkT has good point should someone use some of the advice in this thread for inductive heating.
But just to be clear, incandescent lighting is a resistive load for all practical purposes when used with 50-60 cycle AC. So Zero-Voltage crossing switching techniques also result in near Zero-Current switching for the type application described in this thread.
For one explanation as to why see: