Shop OBEX P1 Docs P2 Docs Learn Events
Of projects and SSRs — Parallax Forums

Of projects and SSRs

henrytjhenrytj Posts: 90
edited 2013-08-30 19:04 in General Discussion
4 months ago I posted some questions about controlling movie lights with a Basic Stamp. Got some very good help.

Over the last couple months, dodging the chaos in my life, I've been piecing some SSR (solid state relay) outlet boxes together $10 or $15 at a time. (will post a picture when I can).

To keep on a small budget, I bought a couple SSRs on ebay. They are Amico/Fotek distributor/brand. But found this

http://www.homebrewtalk.com/f170/fried-my-ssr-409355/index2.html

And wondering if I should be concerned.

For most of the movie light effects I plan to do, the SSRs will be switched on only briefly and intermittently. Usually only minutes to seconds at a time, like for flickering lighting in a horror movie scene.

So, have I cut corners too much? Is the under $10 SSRs one of those "seems too good to be true", because it is. Again, doing this on an extreme budget. Lost my temp job last month and only last week just dodged having my water service turned off.

Henry
«1

Comments

  • Duane C. JohnsonDuane C. Johnson Posts: 955
    edited 2013-08-18 17:03
    Hi henrytj;

    Since you are switching, I presume, incandescent lamps you must be concerned with what is called "Inrush Current".
    Basically, filament lamps when at room temperature may have as low as 1/5th the resistance as when they are "hot" or having been on for a few seconds. So the inrush current can git pretty high for a few tens of mS.

    It would be prudent to purchase SSRs rated at the inrush current of the lamps.

    BTW, if you were simply cycling the lamp, say every 5 minutes, there would probably be no problem.
    However, you are cycling much faster than that so one should be cautious about inrush.

    Measure the lamps cold resistance and calculate the inrush:
    Voltage / Resistance = Current

    Duane J
  • henrytjhenrytj Posts: 90
    edited 2013-08-18 17:11
    Thank you. Having asked this several times before on a couple forums, this is the first mention of "inrush current." The boxes (photo coming soon) are wired with 14-gauge wiring. The SSRs are rated for 25A. I don't intend to push them past 10A and if using my own CFL or LED bulbs, will probably be flowing under one amp. But, on someone else's movie project may want to be able to power 1KW movie lights. And I'm sure someone will ask. So, what is the answer to the question you just posed? How does one calculate in rush current?

    And for something like lightning effects, they might be flickered at a rate of about 1/10 sec semi-randomly for several seconds. So, how does that affect inrush current?
  • henrytjhenrytj Posts: 90
    edited 2013-08-18 17:18
    Here are the pictures of the
    IMG_1734-Sml.jpg
    IMG_1737-Sml.jpg
    IMG_1738-Sml.jpg


    One is my first build and looks very makeshift. Another is my secord build. Still requires a battery for the DC side of the SSR. THe last is my most recent build. Powered by an internally wired AC adapter that powers the DC side of the SSR. As the store doesn't have suitable 4-gang cover plates, a portion of the box is open. But, there are no uncovered AC contacts withing reach of the opening. So yes, still makeshift. The switch bypasses the SSR.
    964 x 842 - 99K
    896 x 810 - 93K
    1024 x 671 - 61K
  • Duane C. JohnsonDuane C. Johnson Posts: 955
    edited 2013-08-18 17:58
    Hi henrytj;

    Technically, LEDs don't have inrush current.
    CFLs, have only a small inrush current. Actually their current increases after several minutes after warmup but not really the same thing.
    Worse yet, CFLs are totally not suitable for fast switching.

    You can measure inrush current with an oscilloscope and current probe.
    Its just simpler to use a meter and measure the lamp resistance at room temperature then just calculate the current.
    (AC Voltage) / (Resistance of cold lamp) = Inrush Current

    Duane J
  • henrytjhenrytj Posts: 90
    edited 2013-08-18 18:49
    Thanks.

    I figured that about LED. Different physics going on there.

    About CFL, well you work with what you got.

    For the movie lights, I don't have any readily available to me. Next time I'm on a movie set I'll see what I can do about measuring the resistance of a cold bulb, if they'll let me. The students are responsible for the equipment that they rent from the film school equipment office. I guess I could call someone at one of the film schools. Not being one of their students (was 30 years ago, but not now) they might not be interested in helping.

    As for oscilloscope. I have one as a prop (might or might not work), but somehow doubts that helps.

    I know I'm not doing everything as right as can possibly be done. If I had the resources, I would. Just trying to do the most, in the safest way, with what I have. I've seen some rather uncertain things on student movies. like one of the students holding the switch to the light in their hand, and flipping the toggle back and forth as fast as they can. (to simulate lightning.) I think I can do better, and safer with my boxes.

    And don't even get my started on the one time I tried to alert the director to a power cord that you could actually smell the hot insulation which was too hot to touch. They got mad at me. Never asked me back on their production. I was being "a problem."
  • henrytjhenrytj Posts: 90
    edited 2013-08-18 18:51
    Anyway, the original question was about the Chinese/Hong Kong sold/made SSRs. Anyone got anything to add about that?

    Henry
  • Mike GreenMike Green Posts: 23,101
    edited 2013-08-18 20:01
    Part of the problem is that, when you buy off eBay or otherwise direct from a vendor in Hong Kong or Taiwan (for example) that you don't know well, you get what you pay for and the vendor is not going to stand behind their products. You don't know whether you're getting the real thing or a knock-off. When you buy from a known manufacturer or distributor, you're paying for them to do the screening of the actual vendor and they'll stand behind the product. Even so, occasionally knock-offs will surface. SparkFun had a blog article once about a part that had a close, but not quite real part marking that got caught in their intake screening.
  • henrytjhenrytj Posts: 90
    edited 2013-08-19 01:30
    Thanks Mike. I agree. But at $30-$40 for new US manufacture, I can't really afford that right now. So it's skimp or not at all. If my luck changes for the better (been waiting years for that) I could always get better SSRs and go back and swap out the others. So right now it's scavenge what I can. Wish it weren't so, but it is. I'll just have to keep a watch for any genie smoke coming out of my magic boxes. I just wanted to see if anyone else has had problems with these. So far in a search I have found only one report of problem with the cheap SSR melting down. It's guerrilla filmmaking. It reminds me of the old WWII movie Operation Petticoat with the pink submarine. They couldn't find enough parts and paint, so they had to do with what they had. They ended up with red and white (pink) paint for the submarine. And they were stealing plumbing from the navy base commander's office.
  • User NameUser Name Posts: 1,451
    edited 2013-08-19 06:50
    Henry,

    A few years ago I purchased some Chinese SSRs that have performed perfectly well. The brand was something like 111AGER. They were rated for 480 V @ 25 A. For cooling I mounted them to inexpensive 486 CPU heatsinks from Jameco (using heatsink compound). I switch them on an off continually (about once a second) w/o any problems at all. The loads in question are either Kanthal A1 wire or SiC 'Globar' heating elements. IIRC, the SSRs cost only $7. I've been really happy with them. Like you, I though domestic SSRs seemed a bit pricey, although buying them used (pulled off old equipment) saved some money.

    BTW: This isn't suitable for everyone, but I've saved even more money by using bare triacs and driving them with a Darlington optocoupler. In fact that is my standard operating procedure now: Propeller-to-opto-to-triac-to-load.
  • tonyp12tonyp12 Posts: 1,951
    edited 2013-08-19 09:36
    LED light bulbs don't use much wattage, are they like 30watt each? (=less than 0.5 Amp)
    Here is one I made for Robert F. it have the same footprint as a Spinneret.
    SpinSSR.jpg?psid=1
    8 SSR's (http://www.mouser.com/ProductDetail/IXYS/CPC1966B/?qs=%2Fha2pyFaduh%252baIUPsSMlqEJTZNx33xahx99Yfbpq3iI%3D) that can handle 3A each (at 115 AC = 345watt)
    And also a mains detection that will show the 60/50hz that goes to the mcu input that you pull-up with 10k.

    I can assemble a board and sell you one.
  • Duane C. JohnsonDuane C. Johnson Posts: 955
    edited 2013-08-19 10:16
    User Name wrote: »
    BTW: This isn't suitable for everyone, but I've saved even more money by using bare triacs and driving them with a Darlington optocoupler. In fact that is my standard operating procedure now: Propeller-to-opto-to-triac-to-load.
    Yup, I did something similar with 8 TRIACs and opto-couplers.
    I didn't use heatsinks at all as this was for strings of Christmas Tree Lights.
    I made 2 of these so I could drive 16 strings.
    A kind of holiday light show.

    Duane J
  • skylightskylight Posts: 1,915
    edited 2013-08-19 10:46
    To minimise inrush current to incandescent lamps a method called pre-heat can be used where a filament is kept warm by being slightly driven not enough to make it bright but just enough to keep warm and minimise inrush.
  • henrytjhenrytj Posts: 90
    edited 2013-08-19 16:42
    skylight wrote: »
    To minimise inrush current to incandescent lamps a method called pre-heat can be used where a filament is kept warm by being slightly driven not enough to make it bright but just enough to keep warm and minimise inrush.

    On my boxes, I have a "by-pass" switch that turns the outlet ON separate from the SSR.

    BTW, I measured the resistance of one of my 500W quartz works lights. It was 8-ohms, cold. So using ohms law that means at 120v it will flow 15-amps. The SSR is rated for 25-amps, I'm guessing that I shouldn't run more that 1000W of light off any one box.

    H.
  • henrytjhenrytj Posts: 90
    edited 2013-08-19 16:48
    tonyp12 wrote: »
    LED light bulbs don't use much wattage, are they like 30watt each? (=less than 0.5 Amp)
    Here is one I made for Robert F. it have the same footprint as a Spinneret.

    8 SSR's (http://www.mouser.com/ProductDetail/IXYS/CPC1966B/?qs=%2Fha2pyFaduh%2baIUPsSMlqEJTZNx33xahx99Yfbpq3iI%3D) that can handle 3A each (at 115 AC = 345watt)
    And also a mains detection that will show the 60/50hz that goes to the mcu input that you pull-up with 10k.

    I can assemble a board and sell you one.

    Thank you Tony. It's a kind offer. But I want to be ready for anything on a student movie set from my own LED/CFL to a 600W movie light that the student might have. Or, maybe even a 1KW. So my boxes, one outlet, one SSR, one channel per box have some nice simplicity. And the lights could be spaced 10 to 30 feet apart. Maybe more. And there's the satisfaction of building them myself. But thanks again. (And the other thing. I don't have any money at the moment. At least until I find another job.)
  • RforbesRforbes Posts: 281
    edited 2013-08-19 19:04
    Tony did an excellent job with the spinneret ssr board. If you can get the funds set aside it's worth it! The board is perfect for driving the ssr as a pilot relay. This way, you can drive power relays withoads up to 40 amps or so if you need to.
  • phatallicaphatallica Posts: 64
    edited 2013-08-19 19:10
    Two things to consider for in-rush:
    1) TRIACs (or SSRs) are designed to handle these surges. Look at the BTA12 datasheet ... even though it is rated for up to 12Arms, it will also handle a one-cycle surge of up to 120A.
    2) Using a zero-cross opto-TRIAC (such as MOC6063) will help to reduce peak stresses. The worst-case scenario is turning on cold, at the middle of the AC cycle. The zero-cross turn-on prevents this, allowing the filament to warm up as the voltage increases.

    The circuit described above has been used by many of the folks who do Christmas light choreography. I use it for general purpose Halloween switching.
  • tonyp12tonyp12 Posts: 1,951
    edited 2013-08-20 08:49
    >own LED/CFL to a 600W movie light that the student might have. Or, maybe even a 1KW.
    One device that can handle 30watt to 1000watt, means that the 30watt situations will pay the same material cost as the high wattage one.
    Sometimes you just have to create two versions.
    I can sell you ($20) a dual 3A that you put inside a power strip if you like to do it yourself, it uses CAT3 cable so it can be controlled 1-50feet away
    attachment.php?attachmentid=103262&stc=1&thumb=1&d=1376092751
    http://forums.parallax.com/showthread.php/149577-Dual-SSR-inside-a-PowerStrip
  • henrytjhenrytj Posts: 90
    edited 2013-08-20 11:00
    Do all SSRs sense and switch at the zero point? Is this a general SSR feature? Or do I need to determine if my Fotek ones do this, and if so, how?
  • phatallicaphatallica Posts: 64
    edited 2013-08-20 13:07
    henrytj wrote: »
    Do all SSRs sense and switch at the zero point? Is this a general SSR feature? Or do I need to determine if my Fotek ones do this, and if so, how?

    If it is this one, then it includes the zero-cross triggering circuitry. If it is something different, then we would need a datasheet, part number, or picture to work with.
  • evanhevanh Posts: 15,942
    edited 2013-08-21 06:55
    henrytj wrote: »
    Do all SSRs sense and switch at the zero point? Is this a general SSR feature?

    Yes, zero-crossing is the norm but not a guarantee. It's called a phase-angle or dimmer switch if it's not zero-crossing.

    AC SSRs do have a habit of failing rather easy, known brand or not. Even though the internal triac has a surge rating well over it's package rating they never live up to it. Always de-rate AC SSRs.

    BTW, There is also DC SSRs. The fancy ones use a photo-MOSFET arrangement known as a bilateral switch, which can handle low voltage AC as well.
  • henrytjhenrytj Posts: 90
    edited 2013-08-21 22:57
    phatallica wrote: »
    If it is this one, then it includes the zero-cross triggering circuitry. If it is something different, then we would need a datasheet, part number, or picture to work with.

    It is this one
    http://www.amazon.com/dp/B00A161UCK/ref=dra_a_cs_lb_hn_it_P200_100?tag=dradis-20
    Or identical in appearance, as best as I can tell.
  • henrytjhenrytj Posts: 90
    edited 2013-08-21 23:31
    Is it possible that these under-priced Fotek SSRs are "dropped" items. There's probably a better term for it. What I mean is this. Some 40... er, um. I mean a mere 15 years ago when I was a lad, my brother worked at a place that assembled electronics. They had a rule, if a component was dropped, it had to be discarded. So my brother, knowing my hobby interests, collected a box of TTL IC chips (flip-flops, counters, and/or/not gates, etc.) for use in my hobby experiments. So, is it possible that these are just such dropped items that are probably good, but the company won't takes chances on them in their primary use, and thus sold on ebay/Amazon, etc? Or, does someone know more about the condition and source of these items?
  • Clock LoopClock Loop Posts: 2,069
    edited 2013-08-22 02:32
    IXYS and other manufacturers don't make it seem like heat is a big deal based on their data sheets(because dealing with heat in products is a large portion of the material costs), but when dealing with many amps at mains voltage, you can bet you will be generating heat with any solid state device.
    Lately the trend is to package ic's in packages that have no heatsink solderpoint, they just expect you to make one with the pcb, and then sandwich the IC&PCB with a custom made heatsink and or fan combo.

    SSR's seem to fit into this category (along with cpu's, regulators, etc... , perhaps using generic heat spreaders on the SSR ICS, thermally bonded to them would help the lifespans, greatly.


    Try pulling 200ma through the prop, continuously. Feel it after 1 hour. DIP, vs QFP44... etc... PUT that in a box, put that with some other heat generating machinery, and it won't live long.

    But if you thermal expoxy'd a heatsink to the IC (even the prop) its lifetime, and even current capabilities will increase.

    Thats another thing, have the PROP overclockers tested IC temp or tried active sink cooling?
  • phatallicaphatallica Posts: 64
    edited 2013-08-24 09:39
    henrytj wrote: »
    It is this one
    http://www.amazon.com/dp/B00A161UCK/ref=dra_a_cs_lb_hn_it_P200_100?tag=dradis-20
    Or identical in appearance, as best as I can tell.
    If your part has the same part number as the one on your link, then it has the same data as was in my link, which indicates zero-cross triggering circuitry.
  • phatallicaphatallica Posts: 64
    edited 2013-08-24 09:54
    henrytj wrote: »
    And wondering if I should be concerned.
    Going back to the original question ... there is a spectrum of reasons for inexpensive power electronics. These range from "grey market" parts - maybe taken out of inventory from a company that went bankrupt - to outright counterfeits. Once it is packaged, there is no simple and guaranteed way to tell. Although, I did once notice our company name incorrectly laser etched on a counterfeit part, most counterfeiters are better than that.

    A failure would either cause a short circuit (ON) or open circuit (OFF), neither of which is a catastrophe when the load is a light. So proceed, albeit with caution.
  • henrytjhenrytj Posts: 90
    edited 2013-08-24 23:55
    phatallica wrote: »
    Going back to the original question ... there is a spectrum of reasons for inexpensive power electronics. These range from "grey market" parts - maybe taken out of inventory from a company that went bankrupt - to outright counterfeits. Once it is packaged, there is no simple and guaranteed way to tell. Although, I did once notice our company name incorrectly laser etched on a counterfeit part, most counterfeiters are better than that.

    A failure would either cause a short circuit (ON) or open circuit (OFF), neither of which is a catastrophe when the load is a light. So proceed, albeit with caution.

    Thanks. I will be keeping an eye on them. A very good reason for not loaning any out, and some might ask. I know a little better now to have a reason to use them sparingly. Not to overload them. Keep the switching of the SSR to the intended effect. And one of the reason for using an outlet with a switch wired to bypass the SSR, so that if a light needs to stay on for a while. It's just a flick of the switch to bypass the SSR, rather than run current through it.
  • henrytjhenrytj Posts: 90
    edited 2013-08-30 06:41
    A curious thing I noticed with the two different types of SSRs I have. One type, when I turn it off, the light blinks on and off barely noticeably for several seconds, getting dimmer with each blink. The other type of SSR I have, doesn't do that at all. So, what causes that subtle difference?

    H.
  • henrytjhenrytj Posts: 90
    edited 2013-08-30 07:07
    henrytj wrote: »
    A curious thing I noticed with the two different types of SSRs I have. One type, when I turn it off, the light blinks on and off barely noticeably for several seconds, getting dimmer with each blink. The other type of SSR I have, doesn't do that at all. So, what causes that subtle difference?

    H.

    Did another search and found two other links describing what seems like the same thing.

    http://forums.parallax.com/showthread.php/138268-Problem-with-LED-light-bulb-and-solid-state-relay

    http://www.electronicspoint.com/solid-state-relay-leak-t6756.html

    H.
  • phatallicaphatallica Posts: 64
    edited 2013-08-30 09:09
    henrytj wrote: »
    A curious thing I noticed with the two different types of SSRs I have. One type, when I turn it off, the light blinks on and off barely noticeably for several seconds, getting dimmer with each blink. The other type of SSR I have, doesn't do that at all. So, what causes that subtle difference?

    There are 3 things missing that are required to provide a reasonable answer:
    1) What is the "other" SSR? You previously advised Fotek as being one type.
    2) What type of light is being used? Incandescent? LED (if so what type of current source)?
    3) Which SSR behaves and which blinks?

    Generally speaking, an AC SSR will have some issues with light sources that require a ballast or driver circuit. Knowing the SSR characteristics (datasheet) and the details of the lighting load would be necessary to offer any better reply.
  • henrytjhenrytj Posts: 90
    edited 2013-08-30 15:04
    phatallica wrote: »
    There are 3 things missing that are required to provide a reasonable answer:
    1) What is the "other" SSR? You previously advised Fotek as being one type.
    2) What type of light is being used? Incandescent? LED (if so what type of current source)?
    3) Which SSR behaves and which blinks?

    Generally speaking, an AC SSR will have some issues with light sources that require a ballast or driver circuit. Knowing the SSR characteristics (datasheet) and the details of the lighting load would be necessary to offer any better reply.

    The other SSRs are older-looking Crydom brand (CL2623, that were given to me) that have a higher 12v-32v DC switching side. For those I use (2) 9v batteries in series.

    What kind of light? Anything I can find. So far have tested with CFL and LED bulbs. Both flash dimmly for a few seconds with the Crydom. I don't know what I might find myself connecting to on someone else's movie set.

    THe Crydom SSRs flash for a few moments. The Fotek do not.

    The other articles I found talk about SSR leakage with low current loads, and the other discussion is about SSRS with/without "snubber" features. I have yet to find a data sheet for the Fotek SSRs, so I don't know what it has.
Sign In or Register to comment.