--
That\'s certainly true for your circuit, but not necessarily for mine
since, without a capacitor, your circuit is much more likely to
drive the coil discontinuously during the long wait for the rise of
a new edge after the fall of the current one.
John Popelish pointed that out by noting that the time constant of
the coils is too short to keep the contacts engaged between cycles
with one half-cycle missing, and that was also borne out by my
experiment, where your method failed 4 out of five times using a
random selection of relays.
To do it safely, there are some concerns about the '12V' relays; if they are intended for auto use, for instance, the application of 120VAC may not be wise.
Why not use the canned solution, an X-10 appliance module and suitable remote controllers? It has the advantage of UL approval. In case of mishap your insurance provider might find that interesting.
On Mon, 18 Feb 2008 17:20:45 -0800, John Larkin wrote:
--- Of course. With your proclivity for argumentative generalization and the ol' bob and weave I sometimes find it necessary to use experimentally derived real-world data in order to pin you down.
---
--- Sorry, that wasn't my intention. ;)
---
--- Better yet, look at this:
Version 4 SHEET 1 960 932 WIRE -1168 -48 -1248 -48 WIRE -1024 -48 -1088 -48 WIRE -704 -48 -1024 -48 WIRE -240 -48 -464 -48 WIRE 176 -48 -240 -48 WIRE -1024 -16 -1024 -48 WIRE -704 -16 -704 -48 WIRE -240 -16 -240 -48 WIRE 176 -16 176 -48 WIRE -1024 128 -1024 48 WIRE -1024 128 -1088 128 WIRE -960 128 -1024 128 WIRE -848 128 -880 128 WIRE -704 128 -704 48 WIRE -704 128 -768 128 WIRE -240 128 -240 48 WIRE -240 128 -304 128 WIRE -192 128 -240 128 WIRE -64 128 -112 128 WIRE 64 128 16 128 WIRE 176 128 176 48 WIRE 176 128 144 128 WIRE -1088 176 -1088 128 WIRE -304 176 -304 128 WIRE -1248 240 -1248 -48 WIRE -1024 240 -1024 128 WIRE -704 240 -704 128 WIRE -464 240 -464 -48 WIRE -240 240 -240 128 WIRE 176 240 176 128 WIRE -1248 368 -1248 320 WIRE -1024 368 -1024 304 WIRE -1024 368 -1248 368 WIRE -704 368 -704 304 WIRE -704 368 -1024 368 WIRE -464 368 -464 320 WIRE -240 368 -240 304 WIRE -240 368 -464 368 WIRE 176 368 176 304 WIRE 176 368 -240 368 FLAG -304 176 0 FLAG -1088 176 0 SYMBOL ind -80 144 R270 WINDOW 0 32 56 VTop 0 WINDOW 3 5 56 VBottom 0 SYMATTR InstName L1 SYMATTR Value .148 SYMBOL voltage -464 224 R0 WINDOW 3 24 104 Invisible 0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR Value SINE(0 34 60) SYMATTR InstName V1 SYMBOL diode 160 -16 R0 WINDOW 0 -45 31 Left 0 WINDOW 3 -109 -7 Left 0 SYMATTR InstName D1 SYMATTR Value MURS120 SYMBOL res -96 112 R90 WINDOW 0 -38 58 VBottom 0 WINDOW 3 -34 59 VTop 0 SYMATTR InstName R1 SYMATTR Value 50 SYMBOL diode 192 304 R180 WINDOW 0 47 34 Left 0 WINDOW 3 23 -2 Left 0 SYMATTR InstName D2 SYMATTR Value MURS120 SYMBOL diode -224 48 R180 WINDOW 0 -44 32 Left 0 WINDOW 3 -107 64 Left 0 SYMATTR InstName D3 SYMATTR Value MURS120 SYMBOL diode -256 240 R0 WINDOW 0 39 34 Left 0 SYMATTR InstName D4 SYMATTR Value MURS120 SYMBOL res 160 112 R90 WINDOW 0 -38 58 VBottom 0 WINDOW 3 -34 59 VTop 0 SYMATTR InstName R4 SYMATTR Value 50 SYMBOL ind -864 144 R270 WINDOW 0 32 56 VTop 0 WINDOW 3 5 56 VBottom 0 SYMATTR InstName L3 SYMATTR Value .148 SYMBOL voltage -1248 224 R0 WINDOW 3 24 104 Invisible 0 WINDOW 123 0 0 Left 0 WINDOW 39 0 0 Left 0 SYMATTR Value SINE(0 34 60) SYMATTR InstName V3 SYMBOL diode -720 -16 R0 WINDOW 0 -45 31 Left 0 WINDOW 3 -109 -7 Left 0 SYMATTR InstName D5 SYMATTR Value MURS120 SYMBOL res -864 112 R90 WINDOW 0 -38 58 VBottom 0 WINDOW 3 -34 59 VTop 0 SYMATTR InstName R5 SYMATTR Value 50 SYMBOL diode -688 304 R180 WINDOW 0 47 34 Left 0 WINDOW 3 23 -2 Left 0 SYMATTR InstName D7 SYMATTR Value MURS120 SYMBOL diode -1008 48 R180 WINDOW 0 -44 32 Left 0 WINDOW 3 -107 64 Left 0 SYMATTR InstName D8 SYMATTR Value MURS120 SYMBOL diode -1040 240 R0 WINDOW 0 39 34 Left 0 SYMATTR InstName D9 SYMATTR Value MURS120 SYMBOL res -1072 -64 R90 WINDOW 0 -38 58 VBottom 0 WINDOW 3 -34 59 VTop 0 SYMATTR InstName R7 SYMATTR Value 50 TEXT -1248 400 Left 0 !.tran .05 uic
Note that either way (with the 50 ohm resistor internal or external to the bridge) clamps the coil to a couple of diode drops _below_ ground, so I still don't know what you're talking about.
However, there _does_ seem to be an advantage to using the external resistor.
The difference is in the ripple current in the coil. The current difference is modest for the L/R values in your model, especially with two diode drops. One interesting thing about relays is that the inductance increases, often by a huge factor, between the de-energized and pulled-in state [1]. Which L did you use in your sim? [2]
Dunno. Tell us.
John
[1] which gives power AC relays a huge advantage.
[2] It would be awful to do a fully time-accurate model of a relay pulling in, what with L being a function of the varying magnetic loop reluctance. Ugh.
--
Come on John, knock off the bullshit.
You wrote:
"Better would be to put the resistor *before* the bridge, so
diodes clamp the coil voltage close to zero."
Which states, as I read it, that without the resistor being placed
before the bridge the coil voltage wouldn\'t be clamped close to zero
volts.
The truth is that either configuration clamps the coil voltage to a
couple of diode drops _below_ zero volts, so your premise is flawed.
No BS. LT Spice lets you see currents. Look at the currents in R5 and R1. The R1 current is decaying faster. The advantage will be more pronounced when the coil L is higher, as it will be for bigger relays and/or when when the armature is seated.
What I said wasn't a "premise", it was a fact. If there's current in the coil, and there's a resistor directly in series with the coil, there must be voltage drop in the resistor, so that changes the voltage in the coil. L/R drops in half if you double R, which is exactly what happens if you put the resistor after the bridge.
What's being clamped is the voltage across the coil+resistor series pair, which is not the voltage across the coil.
Well, coil L increases a lot when the armature pulls in and closes up the air gap.
Thanks! That's exactly the point I've been making.
How so?
Over DC-coil relays, especially in demanding industrial applications. You can design an AC-coil relay to have a huge initial coil current, to really slam the contacts closed against a powerful spring. When the armature fully seats, inductance increases, coil current drops, and the thing stays nicely latched at low coil current. A DC relay with similar mechanics would fry the coil, because current is determined by the copper resistance and doesn't drop when the armature seats.
I have seen DC relays and solenoids with auxiliary contacts that simulated this effect.
Whit3rd, thank you for your reply. I'm sorry for the delay in my response; I let life run me a bit this week and I hadn't been able to get the materials I needed, until yesterday, to test some of the other suggestions so I hadn't been following this closely.
I have not found an answer to the question of whether or not the auto relays can handle the 120VAC, and without that answer I will not use them for safety reasons, as you mention. I do have ones that are rated for 120VAC and 250VAC and 300VAC, so I still have something to play with.
I looked up the X-10 stuff, thank you. What I see is that the devices are mostly used in connection with other X-10 devices whereas what I'm looking for is automation; can I control this, ultimately, with my computer? What I'm envisioning here is that the devices will be installed at my ceiling lights and other permanent power devices in my home and controlled via low-voltage switches, including ones that can be driven by my computer (via the parallel port (I have built these simple devices along the lines of the simple Kit 74 that somebody makes), or USB).
I have found some good information through testing of the suggestions made to my posting that I will summarize in one post in reply to my original post to centralize the information for anyone who wants it.
Hey, all, thank you for your replies and information. I am sorry for the delay my response; I let life run me this week and only got to my electronics shop yesterday.
I had several suggestions on how to solve this and I'll go over them here; one to provide feedback to the replies and two, to offer this information to maybe help others who come along and read it with the same question I had.
First, John Popelish suggested running the relays with 12VDC and that the line-loss would be negligible. I did the math on the line-loss, which I had not conceptualized or understood correctly before his post and it seemed like a good idea. I had some crappy wire already run out (I took a headphone jack to RCA Y-cable and cut it in two and spliced some thin wire between the pieces to make a long-distance auxiliary input for the stereo in my shop to run off the stereo in my house). The cable is probably thinner than what John and I had bounced back and forth about and the RCA audio stuff is pitifully thin. The overall cable length is about 80 feet. I ran 12VDC across the cable to a Shrack RP510012 relay with a coil resistance of 327 ohms. The coil snapped briskly in and out when the voltage was applied. John was absolutely correct, the line-loss was insignificant for the resistance of the coil. As I told John I would, I checked the cut-in voltage of the relay. I started with it hooked up and engaged at 12VDC and then turned the voltage down until it dropped out. I then removed the relay and checked the voltage. Cut-out voltage was
2.79 VDC. I then hooked the relay back up and turned the voltage up until it engaged and then disconnected and checked the voltage. Cut- in was very soft at 6.5 VDC. I upped the voltage until I had crisp cut-in which I got at 6.9 VDC. There seems to be no problem with a very long-distance run to control these relays. Thank you, John Popelish.
Next was John Larkin who suggested running the relays directly off 24 VAC with a couple of diodes. I used his diagram and two 1N4004 diodes. One diode came from the AC input to the relay, one went across the relay coil. On the Shrack RP 50012 with a 327 ohm coil and my 24 VAC source (RMS meter says it's 26.9 VAC) the relay chattered constantly. I tried a Zettler AZ8-1CH-12D with a 312 ohm coil. It also chattered. There seems to be some discussion as to why this would or would not work and, honestly, it's beyond my current skill level with electronics. But, FWIW, that's what I tried and that's what I got. Thank you, John Larkin.
Next, Ed suggested I try something involving a Zener and a transistor. I formulated my own idea based on this and then, in response to that, Ed suggested a schematic. It tried his way and mine. His way was to take the 24 VAC (which I full-wave rectified and then buffered with a 2,200 uF 63V electrolytic) and make a voltage divider with two 1k ohm resistors to drive a transistor. I hooked up a LM78L12 exactly as (and the app notes) said, and the voltage divider exactly as he said. I ran the voltage divider output to the base of a
2N4401 transistor. Applying 24VAC (same power supply as used above) gave me a crisp turn-on of the relay and a crisp turn-off. However, the turn-off came about 1.5 seconds after I disconnected the power from the circuit indicating that the capacitor was still carrying the circuit for a little bit, but at least it was a crisp turn-off. I only tried this with the Schrack relay. I then tried the circuit the way I had postulated in response to Ed's suggestion before he gave me the schematic. I did it exactly as he suggested except that I ran the rectified and buffered 24 VAC (about 35 VDC) trough a 30V Zener to the base of the 2N4401. I got the same crisp on and off but the off came much more quickly after I disconnected the power from the circuit. I did the math on the current through the relay coil (12 / 327) and got that it should be 36.7 mA. I measured it in actual use and got about
35 mA. In either case, the current should be well below the maximum current capability of the LM78L12 (100 mA, 140 mA peak), however, the voltage regulator was getting very, very hot (enough it was painful to touch and I could smell the chip). I checked the actual current through the voltage regulator to the coil of the relay (which is where I got the number above; 35 mA). At 12 volts that should be about 0.4 watts, which I thought the LM78L12 could handle but the heat seems excessive. I may need to put a heat sink on it. Regardless, the idea works. Thank you, Ed.
Thank you all for your help.
The best, I think, is going to be to use the 12 volts DC from the beginning as it is simpler and seems to work very well. With what I think I understand now, I should be able to easily calculate the maximum run of wire for the 12 VDC so it should be easy to know if whatever run I choose will work or not. If, for some reason, that doesn't work for the distance I might want to run, then I can switch to the 24 VAC with the voltage regulator/transistor/zener.
Thanks for adding your conclusions to the thread. It will make it more useful to those who eventually find it, in a search to solve similar problems. I'm glad you have gotten useful answers and added to your understanding.
Nice! Thanks for letting us know what you did and the results you got. :-)
Measuring the pull in and drop out voltages can be an eye opener - amazing to many that a 12 volt relay is still energized at 3 volts, until that observation prompts some thinking.
The heat you noticed in the 7812 can also be an eye opener. Some people think "I'm using it a well below the max, whyinthehell is it getting hot". You went the right way and figured the watts. The 7812 will handle that .4 watts, but you need to take the heat away for it to do so.
The time delay you found is expected, and a good observation to post as you did. You can experiment to get rid of that and learn something about time constant, if you do not already know it.
There's a lot of people who can benefit from your example of investigating various alternatives, if they do the same.
Thanks again for posting your results. Often people ask a question, get answers, but don't follow up to say how they made out. Nice to hear when someone was as successful as you were, :-)
John, I'm very sorry, the omission was not intentional. Thank you for your input. You were correct, the relays did chatter. I chose not to use the series resistor because it seems something about it runs contrary to my nature (I'm not even sure how to put that into words better). I like the idea of using two relays in series with the load in parallel, but the loads I intend to run won't need that kind of current. :-/
Again, I'm sorry. Thank you for your time and help.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.