Making an X-10 lamp module immune to dimming

Anyone know of a mod to make an X-10 lamp module immune to dimming? If not, does anyone have any suggestions on how to accomplish it? I have lots of extra modules to experiment on, and it would seem that there's a control line that tells the triac how much to "chop" that could be modded to always run full tilt. That way, a brighten or dim command wouldn't affect them.

I found the schematic at Dr. C's site:

formatting link
He also did some work attempting to make the switch smarter

formatting link
by dropping in his own PIC, but was overtaken by events and commercial units becoming available that did what he wanted.

It looks like the triac, BTA10-400C, gets controlled by pins 6 and 8 of the

78561 microcontroller. The trick is to make it deaf to only bright and dim, not on and off.

-- Bobby G.

Reply to
Robert Green
Loading thread data ...

As opposed to just using an appliance module?

Reply to
Bill Kearney

Yes, for quite a few reasons.

1) Most importantly they do not clack

2) I have about 40 of them doing nothing that I would like to put in service

3) they can be used for some CFL's as long as they aren't dimmed

4) I seem to have less of a flashing and phantom restart problem using them with CFL's.

5) They respond to ALL LIGHTS ON and OFF, appliance modules don't

-- Bobby G.

Reply to
Robert Green

The instructions for the mod you seek can be found here:

formatting link
down on the left to "Lamp Modules" and the sub heading "Making a Silent Appliance Module" and start warming up your soldering iron!

Ido's site is THE repository for X10 mods!

Robert Green wrote:

Reply to
BruceR

IC pin 8 is the output that drives the triac. The delay of that pulse with respect to powerline zero crossings determine the brightness. It is not clear how they adjust the phase, but it is probably done by some sort of timer in the IC whose count for each half cycle is a function of the dim level.

Since older lamp modules come on at full brightness, the problem only occurs if they are manually sent dim commands. Either just don?t do that, or put them on their own housecode where they can?t be directly accessed. Then have your Ocelot or HV map manual commands to only ON/OFF on that housecode. The macro can also handle the ?ALL? issues.

We have a bunch of lamp modules that are only used for ON/OFF control. Most power CFLs, but one runs our hot-water recalculation pump. Can you imagine CLACK - - - -CLACK every 15 minutes from an appliance module?

Jeff

Reply to
Jeff Volp

Reply to
BruceR

Thanks! I saw that and it's close but no ceegar. One of the reasons to do this is so that lamps CAN respond to all lights on and off. I suppose in the spirit of "Thomas Edison" whose strategy was often "try anything you can think of" I'll try cutting the control lines from the PIC to the triac to see what happens. I had hoped to avoid using virtual housecodes as Jeff suggested to filter the dim and bright commands. It just seemed to me that the bright and dim control circuit paths might be separate enough from the on off circuitry that one could be neutered without affecting the other. It would take a far smarter person than I to determine where to snip based on circuit analysis, even with Mr. Bloom's helpful schematic. However, reading through Ido's explanation of his mods, I've got a few more clues as to how the module operates.

-- Bobby G.

Reply to
Robert Green

Cue Dave for this anti-CFL rants...

Reply to
Bill Kearney

| > Ido's site is THE repository for X10 mods! | | Thanks! I saw that and it's close but no ceegar. One of the reasons to do | this is so that lamps CAN respond to all lights on and off.

The described modification does not affect the module's response to all lights on. If you actually have lamp modules that respond to all lights off (none of mine do) it won't affect that either.

| I suppose in | the spirit of "Thomas Edison" whose strategy was often "try anything you can | think of" I'll try cutting the control lines from the PIC to the triac to | see what happens.

Nothing useful will happen...

| It just seemed to me that | the bright and dim control circuit paths might be separate enough from the | on off circuitry that one could be neutered without affecting the other.

There is but one path.

Dan Lanciani ddl@danlan.*com

Reply to
Dan Lanciani

Since we have had two seriously smokin' CFL failures that filled the house with acrid melting plastic smoke, I am afraid to leave the "don't dim" rule in the hands of operators or my weak HA programming skills. Another CFL "Burning Man" tribute would have some serious SAF repercussions. It's just too easy to press DIM with too many different types of controllers, from maxi's to Palmpads to keychains when you don't mean to. Putting all the lights on phantom codes would make the HV or CPU-XA a central failure point I'd rather not have given my mediocre programming skills. It's not the end of the world if this isn't doable, but it would be awfully nice to put the huge box of them to use and to get rid of the clacks. I really don't want to go the virtual housecode route, but that seems to be the only option if I can't make the modules themselves deaf to bright and dim commands.

I believe the first test will just be to simply disconnect pin 8 from the control line. Hopefully it will be easy to rebridge if that turns out not to work. Also, hopefully, it won't incinerate itself after snipping that line. I guess I can trot out the old 20mm ammo box I have for containment of explosive experiments - well, maybe just leave it on a fireproof surface to test . . .

That would have very bad SAF. It's good to know I'm not the only one that wants to use lamp modules for CFL's and other uses. I assume you're using a virtual housecode to filter out any "bad presses" that would have resulted in sending a DIM to the pump controller lamp module. I really would like the ALL LIGHTS ON/OFF command to work with CFLs. Also, I don't know whether it's dumb coincidence, but my N:Vision 23W flood does not relight or flash on a lamp module but does when connected to an appliance module. The relighting and flashing is very bothersome. Come to think of it, that's probably why you have very little trouble with relighting and flashing CFL's but I do - you're running them from lamp modules and I have been using appliance modules. The lightbulb over my head has just lit up!

Thanks for the input, Jeff.

-- Bobby G.

Reply to
Robert Green

Whoops, that was a typo before. Pin 8 is used for the local control. Snipping that will only disable local control. Pin 6 drives the triac. That line is essential. If snipped there will be no on/off control at all. The dimming phase control is inside the IC.

It might be possible to extend the pulse coming out of the IC so that if there is ANY pulse at all, the transistor will remain on through the beginning of the next cycle so the unit will switch on right after the zero crossing.

Jeff

Reply to
Jeff Volp

Actually, I think it's pin 6 that drives the triac, via C337 (assuming that schematic is correct).

A while back, I hacked around in those controllers. Here are some of my notes:

"The IC generates a series of pulses, with at least one always present, near but before the zero-crossing (i.e. even at "off" the triac is triggered, just very late in the half-cycle).

"As the "brighten" signal is sent, the pulse first slides earlier in the half-cycle, and then multiple pulses appear, with more and more as full on is approached."

What I wanted was a single duration-modulated pulse (wider for brighter, of course), and here's how I got it:

1) Add a signal diode (1N4148) in series with the signal from IC pin 6 (i.e. cut the trace and hook it between the two points "B" on the schematic, cathode towards the transistor) 2) Change the 2.2 k resistor to 12 k. 3) Connect a 0.1 mfd. capacitor (ceramic is fine) from the cathode of the diode to V- 4) Change the 330 ohm resistor to 2.2 k
Reply to
isw

Your are correct about transposing pins 6 & 8. I caught that myself in an earlier post.

I was thinking along the lines of a diode and cap too. I have not looked at that signal myself. However, if there is always a pulse there, a simple circuit may not be sufficient. A pulse late in the cycle fed through a diode - capacitor network into the transistor would trigger the triac early in the next cycle. Or, does that pulse go away completely when the module is OFF?

Jeff

Reply to
Jeff Volp

How much current does the triac's gate draw? I'd be worried that the module's power supply can't handle leaving the transistor on for long. These supplies are always by design near their limit.

The $1 modification avoids the power problem by getting the gate drive directly from the line/load. The author appears to know what he is doing so I think if there were a simpler way he would have found it. And it's already pretty simple...

Dan Lanciani ddl@danlan.*com

Reply to
Dan Lanciani

Not if you get the time constant right.

According to my notes (several years old), there is always at least one pulse there. The diode and capacitor, plus changes to the values of the resistors, will integrate the pulse train into a single pulse that gets wider as the dimming moves towards full on (the first pulse in the train is always at the time the triac *should* fire; maybe all the others are just "insurance" or maybe it's just the way that chip works. When the unit is at "full off" that single pulse is very narrow, so an RC integrator works pretty well (at least, it worked for me).

-- Side note

I took the optically isolated integrated pulse, and integrated it again to a DC level, then fed it to a National LM-3914 Dot/Bar Display Driver. I was looking for a cheap way to control a lot of off-on devices from a single X-10 unit. I figured that by counting "brighten" pulses, I could make whichever of the 3914 outputs hot that I wanted. Problem was, the X-10 unit was not consistent from one time to the next, so everything worked just fine except that say, 5 pulses, would not always make the same output hot; sometimes the one "before" and sometimes the one "after". Bummer.

Reply to
isw

As I said, and according to my notes, there is *always* at least one pulse there, even if the unit is fully dimmed.

There's already a current limiting resistor there (39 ohms). One could try increasing it, but if the drive current gets too low, the triac may not trigger completely, and could dissipate more heat than would be a good idea. Based on my memory of the pulse train that drives the triac, I'd say that with the circuit I suggested, the integrated gate current is no more than two or three times what it is in the original circuit.

Isaac

Reply to
isw

a

You're correct. At first glance I thought he was talking about putting a solid state relay inside an appliance module to eliminate the clack. After your comment, I went and tested the modules I have and while the wall switches respond to All Lights ON/OFF, the lamp modules don't. They only respond to All Lights ON. Bummer. Well, that makes the project somewhat less attractive. At least the All Lights ON command is more useful to me than All Lights Off would be. How is that the X-10 security system flashes the house lights? Do lights on lamp modules turn on but not off? My memory seems to be that all lights turned on and off, but I could easily be mistaken. I think it's Mad Cow disease setting in.

Well, such is the way of the "grunt and crank" method. I would have at least re-discovered the local sense mod for lamp modules. My whole premise was based on a misunderstanding of how triacs work. I thought they were merely high speed electronic relays.

It's become apparent that the triac can only do its magic when synched to the zero crossing. That became clear as I read through the modification details. Being a non-electrotechie, I assumed that a triac just passed current until the trigger voltage was no longer applied, like a solid state relay. I also thought its other "claim to fame" was how incredibly quickly it could switch and that tiny triggering currents could control much large currents. I just had hoped like the local sense problem, a simple snip would do it. Now it seems like the simplest mod is to extract the ZC data, optically isolate it and then use it to inject the synchronizing signal into the triac, assumable for it be able to reverse its polarity and pass current flowing in the opposite direction for each half cycle of the AC line.

I've just disassembled a lamp module and it looks as though it's not as crowded as I recall. I have enough of these to practice on so that when the weather turns hot, I might just attempt to undertake the modification. I'd still rather just clip a resistor lead to fix them but the discussion has at least educated me as to why that's likely to have no effect.

Perhaps you could be kind enough to explain what would happen if the line to the triac gate was high all the time. Would it conduct only on the first half cycle, or not at all or would the magic smoke escape? (-:

-- Bobby G.

Reply to
Robert Green

As I understand it now, the pulse to the triac has to occur twice every cycle and has to be alternatingly negative and positive to allow each half of the AC cycle to pass. The dimmer does its magic by narrowing the width of that pulse - shorter means less power gets to the device and its dimmer. As I said to Dan, I thought a triac was just a very fast solid state relay.

It sounds like the mod that uses the alternistor and the optoisolator does just that. It seems that the optoisolator extracts the ZC data and then the additional circuitry creates its own full width pulse independent of whatever length pulse comes out of the IC. So if the module was dimming the output, it wouldn't matter because that pulse is driving a latching circuit that stays on until the next ZC, correct?

-- Bobby G.

Reply to
Robert Green

Does that mean that the pulse width does not vary, but that the level of the light is determined by how many pulses of a fixed width are received during an AC cycle? I assume that would me they are obviously much shorter in duration than 1/2 the AC cycle. It sounds like they are "rechopping" the pulse into smaller pulses and that they aggregate to give the desired end level. I am not explaining that well, I fear. It seems like there are two pulse aggregations, a wheel within a wheel Ixion sort of thing.

Why did you want to do this? What is the benefit of going to pulse width from their system of number of pulses per cycle?>

Bear with someone with no experience but at least a little interest in the subject. That's a high speed switching diode, AFAICT, and used in series it will pass current in only one direction . . .

You would normally change a resistor value to change a time value in a dependent RC circuit or to protect some other component from drawing too much current, correct?

OK, I'm lost although after I post this I will probably find out I was lost far early in the process. I'll stop here even though I have some more questions about your other modification suggestions because I'm a NOT a professional engineer who has been doing things like this (or worse) for well over forty years. I also know that my level of competence is NOT sufficient for me to tackle it without lots and lots more discussion and education.

I've already discovered that X-10 gear has naked 110VAC running all over the circuit boards, making tinkering by noobs like me a lot more dicey than trying to solder remote contacts to a coin cell powered keychain controller circuit board. The only way that power source could hurt you is if you ate it or someone shot it at you with a slingshot. More my kind of "futzing." With the lamp module apart in my hands I have to really marvel at all the engineering that went into them and how well that simple circuitry has endured.

-- Bobby G.

Reply to
Robert Green

But it also pretty clearly states that advanced soldering skills are required. )-:

From reading more into that particular mod, I've gleaned that it's necessary to feed the ZC timing to the triac in order to deal with the reversing polarity of the AC signal. That wasn't something that was at first clear in reading up about triacs. Obviously, electronics is not a real hobby of mine. I am only in it this far to try to reclaim some usefulness out of a carton full of WS-467 wallswitches and LM-465 lamp modules and because of a general interest in how this stuff works.

I was hoping that the required mod would be a simple one, like the disabling of local sense on some appliance modules. You just need to open the case and snip the lead to a diode that, IIRC, is conveniently mounted vertically. That I can handle. Piggy-backing chip sockets or adding additional microprocessors is not within my skill set, even with my new Weller soldering station. My work looks somewhat better, but it's still nothing I would trust a 110VAC device with. Also, considering all of these came basically free with Ebay X-10 lots I bought for the other components in the lot. Many of them came nearly free with the infamous "X-10 free vouchers" combo packs. It would probably just be easier to sell them all than butcher them. My box of X-10 patients that didn't survive various complex mods is impressive. )-:

From what I had read here and elsewhere, it sounded very much like triacs were solid state switches that could pass large amounts of current via the input of very much small current. They also appear to be able to switch incredibly quickly compared to mechanical switches. But a practical application like this makes the theory easier to understand. When I read that they were bidirectional, I thought it meant passing AC freely, but it's a lot more complex than that.

-- Bobby G.

Reply to
Robert Green

Cabling-Design.com Forums 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.