Jump to content

LOR board and rectification


Dr. Jones
 Share

Recommended Posts

Has anyone tried rectifing the input to a 16xx or 16PC? Can the board handle 120v of quasi DC?

Reason - by rectifing the outputs, you can minimize the 60 cycle flicker of HW leds. But what about the inputs???

Link to comment
Share on other sites

So far, I don't see why the left side of a PC series wouldn't be OK with it, but I would expect problems with the transformer on the power supply on the right side..

Of course, not all half wave LED strings can handle double the on time.. On top of that, many strings have each half wired in reverse of the other half...

Link to comment
Share on other sites

Well I for one see a problem if not taking in account the step down transformer that is for the power supply that powers the logic circuits. How do you expect for the triac to shut off? A triac is a double scr with one half in reverse to the other half. What I would see is that one half would never turn on and the other half, once given bias to turn on, would never turn off again till the power is removed from the board (aka the plug pulled to the controller).

If you want to rectify the AC voltage, you will need to do this after the the output from the board.

Max

Link to comment
Share on other sites

If you presume that only one SCR being in use should cut the rating in half, the 16A triac becomes an 8A triac.. Give it a 100% safety margin like LOR does normally, and you have 4A.. I don't think most people will exceed that on a channel of LEDs...

For SCRs to shut off, current has to go to zero, or at least less than the hold on current. The voltage does not actually have to invert.. If the LED string and feed wiring has no inductance or reactance, the voltage going to zero between half waves should be enough to allow it to switch off.. However, even a little inductance from the wiring might keep current from dropping below the hold on value.. Slightly offsetting this is that the rectifier should create a 1.4 volt dead band around zero. This should give a measurable duration of zero voltage around the true power line zero crossing, which gives the channel a greater opportunity for current to drop below the hold on current.

As for the card power supply, that only draws from one of the power inlets, so it should not be impacted if only the other card inlet is rectified..

I'm not going to say that it is a good idea, or that there aren't probable issues. Just that the side of the card not powering the logic is probably OK with the rectified, unfiltered AC..

It isn't that difficult to test, but it isn't on my short list of things to do either...

Link to comment
Share on other sites

-klb- wrote:

Of course, not all half wave LED strings can handle double the on time.. On top of that, many strings have each half wired in reverse of the other half...

I bought a few strings of half wave white LED strings. I put a full wave rectifier in front of them, and yes, they were then brighter. Since I was going to install them semi-permanently on my roof, I wanted to make sure they could run this way without failing.

So I plugged a string into an outlet in the garage, and left them lit for 24+ hours. No problem. I figure that in actual use, they will be on intermittently, at lower temperatures than found in the garage, so they should be fine.

By the way, the full wave rectifier I found was in an LED string from ChristmasDepot.com. It supplied rectified unfiltered DC at the female socket at the end on the string, where you would expect to find 120VAC.

Doing it this way was a lot easier than supplying unfiltered DC to the input of a controller, especially since the controller I was using was an 8-channel.
Link to comment
Share on other sites

klb,

you are absolutely correct about the SCR has to go to zero and not reverse voltage. Thanks for covering my error and getting the correct info out there. Now weather or not the gent is going to filter his rectified DC will have an impact on the performance of the triacs. and from your explanation, it looks like he should not filter and hope that the voltage drops off enough between half cycles.

Max

Link to comment
Share on other sites

There is no difference in the LED's used for HW or FW strings. It is just the additional components to make the string FW. Yes the LED's will look a bit brighter as they will be on longer, the average current will go up but the peak current will remain the same.

I do not think all HW strings would work, some are wired that half the string comes on during the positive cycle and the other half comes on during the negative cycle. I use 70 light strings and that is how they are wired, cannot add the Full Wave Rectifier unless I take the secon half of the string and reverse the two AC wires to the second string.

Making the power input unfiltered DC will probably not work reliably also the controller I think is using the 60 hertz frequency for the shimmer and twinkle affect.

Link to comment
Share on other sites

Dennis Cherry wrote:

Yes the LED's will look a bit brighter as they will be on longer, the average current will go up but the peak current will remain the same.

The most common reason for LEDs to fail or degrade is elevated temperature. Keeping them on longer can cause them to run hotter.

Making the power input unfiltered DC will probably not work reliably also the controller I think is using the 60 hertz frequency for the shimmer and twinkle affect.

The controller gets the 60 Hz frequency, which it uses for not only shimmer, but also for ramps and intensity (I believe twinkle is just random on/off), from the zero crossing detector on the right side (channels 9-16) of the controller. Rectifying the input to the right side would probably destroy the transformer. However, the left side would still probably work with unfiltered DC.

It still wouldn't work with the half wave strings you have where half the bulbs light on alternate cycles.
Link to comment
Share on other sites

Max-Paul wrote:

klb,

you are absolutely correct about the SCR has to go to zero and not reverse voltage. Thanks for covering my error and getting the correct info out there. Now weather or not the gent is going to filter his rectified DC will have an impact on the performance of the triacs. and from your explanation, it looks like he should not filter and hope that the voltage drops off enough between half cycles.

Max


Max - No filtering as - you are aware - I want to keep the zero for the scrs.

Yes I would be using the left side (1-8) side of the board as I don't know how the electronics would react.
Link to comment
Share on other sites

Dr. Jones wrote:

Max-Paul wrote:
klb,

you are absolutely correct about the SCR has to go to zero and not reverse voltage. Thanks for covering my error and getting the correct info out there. Now weather or not the gent is going to filter his rectified DC will have an impact on the performance of the triacs. and from your explanation, it looks like he should not filter and hope that the voltage drops off enough between half cycles.

Max


Max - No filtering as - you are aware - I want to keep the zero for the scrs.

Yes I would be using the left side (1-8) side of the board as I don't know how the electronics would react.
Well Dr.
I'll tell you two things. You can not put a transformer in a DC circuit. Filtered to be true DC and I do not know how well it would work with a pulsed DC voltage applied to the primary (people we are not talking about an inverter at this time.So do not confuss this discussion with an inverter). But more than likely it will still burn up, but I am not an EE, nor have I tried it. But enough experience and knowledge to tell me not to try it. See the transformer that comes on the LOR board is wired for impedence not for ohms. For those of you following this discussion that are not familiar with Impedence and ohms. Impedence is a AC measurement of resistance that takes in account the inductance and capasadence of the circuit. Where as Ohms is the resistivity of a circuit even though there is a coil or cap in the circuit. And even if the primary does not burn up. I expect that the voltage out of the secondary might not be compatible to run the logic circuits.

But hey if you got the time and money to try this. We all would be interested in hearing your findings. Just please be sure to fuse everything well and have a panic kill switch near by just in case. And please use protective equipment like a GFI.

Max
Link to comment
Share on other sites

Max

I am only going to attempt on the left side - the only circuitry on the left side are the triacs, a resistor, and the opto. My only concern is the opto.

I know the circuitry on the right side of the board will have major issues being powered with a rippled dc wave. I in no means am going to try this, as I don't have the time or the money to replace a controller.

Link to comment
Share on other sites

If you use a full-wave rectifier, then you will not get a zero-crossing and the triacs will not turn off. Rectifiers are fairly cheap, and my view is that you are far better off doing this to every output you want dc from. You will have your dc, and the triacs will see the zero crossing required to turn them off.

If you want something on the input side, a half-wave rectifier will work as it will give the zero-crossings the triacs need, but you will only have half the power to the LEDs, as you are only using either the positive or negative half-cycles.

Regards,



Alan C.

Link to comment
Share on other sites

Alan C wrote:

If you use a full-wave rectifier, then you will not get a zero-crossing and the triacs will not turn off. Rectifiers are fairly cheap, and my view is that you are far better off doing this to every output you want dc from. You will have your dc, and the triacs will see the zero crossing required to turn them off.
Max: I believe I said the same thing in my first post. And fully agree with you here Alan.
If you want something on the input side, a half-wave rectifier will work as it will give the zero-crossings the triacs need, but you will only have half the power to the LEDs, as you are only using either the positive or negative half-cycles.
Max: Dont think I can agree with you here Alan. A half wave or single diode is still going to block half of the wave. Hence no crossing of the zero line. might of fact the diode will stop conducting .6volts above the zero line and wont start to conduct till .6v above the line.


Regards,



Alan C.

Max
Link to comment
Share on other sites

Max-Paul wrote:

Dont think I can agree with you here Alan. A half wave or single diode is still going to block half of the wave. Hence no crossing of the zero line. might of fact the diode will stop conducting .6volts above the zero line and wont start to conduct till .6v above the line.


Yes to some extent, but it is the output from the diode that is important, not how the diode is operating. With the single diode arrangement, the triac will only be receiving power for half of every cycle. During the other half of the cycle, the triac will receive no supply and (in theory) stop conducting. It is roughly equivalent to unplugging the supply for half of every cycle.

Regards,

Alan C.
Link to comment
Share on other sites

Ok, All you all listen up..

Rectifying AC does not mess up the phase angle. Rectifying without filtration does not get rid of the zero voltage points.. It does result in a loss of zero crossings, that are coincidental with the zero points. Conversationally we talk about triacs switching off at the zero crossings.. But it is not the crossing that causes the switching, but rather the zero current, that with a resistive load is coincident with the zero voltage...

To test this, I finally located a reasonable sized bridge rectifier, with reasonable leads, rather than one of the DIP packaged ones that seemed to be all I could find around the house for a few days.. I also chased down a three wire extension cord to cut up, and a euro style terminal block.. I then installed the rectifier in line in the cord with the terminal strip.

I plugged what I believed to be the right side LOR inlet cord directly into a power strip, and turned it on. This verified that this cord was indeed powering the logic, as feeding rectified power to the transformer is nothing but bad news. I then plugged the other inlet into my modified cord, and plugged the modified extension cord into the power strip..

No issues so far... I took two strings of 100 count Walmart mini lights, and plugged one into channel 1, and another into channel 9.. I fired up the HWU, and had it ramp all channels from 0 to 100 percent over 10 seconds. I could perceive no difference between the two sets..

I located a couple of two year old, unused, 30 LED Walmart LED strings... I plugged one into channel 9 and thought that a seizure might be less unpleasant than looking at them.. I plugged the other set into channel 2, and liked the nice steady output right away.. To confirm that it is indeed a half wave string on a rectified supply, I turned the plug around, and they did not light.. I went ahead and plugged it back in so they would light.. I had the unit ramp from 0 to 100 again. Other than a glitch part way through the ramp, which both sets appear to do, the ramp worked fine. I'd be more positive that both sets had the glitch in the ramp part way through, but the flicker on the half wave set is so bad it pretty much masks the glitch...

I next plugged the incandescent strings into the LED strings, and ramped again. This time, the glitch in the ramp went away.. For my last test, I completely unplugged the incandescents, and ramped, and went back the same behavior as initially.. The fade worked, and both sides had the glitch part way up the ramp...

So, I'll again say that there is not an issue with the LOR boards accepting, switching, and providing effects with rectified AC line power on the side of the card that does not supply the logic.. The disclaimer is that inductive or capacitive loads will cause problems, but then again, they usually have issues on LOR anyway.

Maybe some day I'll get curious enough to take a look at what is going on around the glitch with a scope, but don't count on it...

Link to comment
Share on other sites

An interesting follow up note.. I upgraded the firmware on this PC card from 4.02 to 4.30, and the glitch part way through the ramp up in the HWU has disappeared...

Link to comment
Share on other sites

klb,

Thanks for the report. Ok, I lived and learned something today about the triacs. I would have done this myself, but being a newbie to LOR and still need to solder up my boards. Well I suppose that took me out of the running this test myself at this time.

Again, thanks for the report.



Max

Link to comment
Share on other sites

Some further notes on this..

These particular strings have molded resistor packs in line. On the one running half wave, it was not particularly warm to the touch.. The one running full wave got quite warm to the touch...

One idea for ensuring that you don't accidentally plug the wrong side into the rectifier would be to build these as single inlet controllers, and build the rectifier into the connectors that power the second inlet from the right side inlet...

Link to comment
Share on other sites

They may not be obvious, but I really expect there to be a resistor in there somewhere.. Though in different lengths, there may be much less energy to dissipate from the resistor than in these 30 LED strings..

The reason there really should be a resistor in there is that current through a string of LEDs varies exponentially with voltage, not quasi linearly as with a light bulb...

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
 Share

×
×
  • Create New...