Two Channels Duplicated - Isolated Tab Triacs No Longer Isolated?

[Steven]

The controller is a CTB16PC v1, firmware version 4.32. I'm testing it using the Hardware Utility and modern 100-bulb LED strings.

When I turn on either channel 12 or 13, lights connected to both channels turn on, to the maximum dim level of those 2 channels.

Some investigation with the power off shows there is a low resistance between the hot leads on those channels. Further investigation shows a low resistance (<Ω) between the hot (switched) connection and the tab of the triacs, which are both connected to the heat sink. Since the CTB16PC is mounted in the plastic case, the heat sinks are not grounded. Aside from the channels being shorted to one another, this also means live voltage is present on large metal areas (when the show is running).

I assume this means the triacs for both channels 12 and 13 now have an internal short from T1 to the mounting base. Aren't the triacs used in CTB controllers the isolated tab kind?

Since the triacs seem to work otherwise, should I just treat them as not having an isolated tab, and just get some insulating hardware for them? Or is this just a sign that they will probably fail completely in the future?

 

[TheDucks]

What happens if you turn on port 16?

12 is bits 8 and 4

It may be the (port) decoder

[Steven]

6 minutes ago, TheDucks said:

What happens if you turn on port 16?

I didn't try that port in this test, but when the show was running in December the only issue noticed was ports 12 & 13.

Also, I checked with a meter directly from the tabs that connect to the pigtails, and channels 12 and 13 are the only ones that have nearly 0 resistance to the heat sink, so I'm pretty sure it's a triac problem.

[TheDucks]

1 hour ago, Steven said:

I didn't try that port in this test, but when the show was running in December the only issue noticed was ports 12 & 13.

Also, I checked with a meter directly from the tabs that connect to the pigtails, and channels 12 and 13 are the only ones that have nearly 0 resistance to the heat sink, so I'm pretty sure it's a triac problem.

Maybe your flat insulator (the screw has a nylon bushing as well) is bad on those. The heat sinh should NOT be connected to the Tab (

FWIW, I thought it was grounded. The fuse should have blown... OK I checked my assembly manual and the grounds go only to the clamp bar 🤒

[Orville]

2 minutes ago, TheDucks said:

Maybe your flat insulator (the screw has a nylon bushing as well) is bad on those. The heat sinh should NOT be connected to the Tab (

FWIW, I thought it was grounded. The fuse should have blown... OK I checked my assembly manual and the grounds go only to the clamp bar 🤒

That's odd, on my v1 ctb16pc's the heat sinks ARE connected to the ground bar, as the tab on the triacs in mine are definitely grounded.  I checked with a continuity tester and the heatsinks and the triac tabs connected to them are grounded to the green ground wires connected to the metal bar.  All mine were purchased as RTG Units right out of the box.

So I would think the triacs tabs should be connected to ground.  Just know mine are, and they work, so I'm not going to unground them.

[Steven]

27 minutes ago, TheDucks said:

Maybe your flat insulator (the screw has a nylon bushing as well) is bad on those. The heat sinh should NOT be connected to the Tab (

The assembly manual makes no mention of an insulator or nylon bushing, but simply says to apply heat sink compound between the triac and the heat sink.

It also specifies part number BTA16-600BW for the triac, where the "A" in "BTA" means "insulated" (from tab). I think what happened in my controller is that an internal short inside the triac has turned it from an "insulated" devices to a "non-insulated" device. I guess I should consider myself lucky that it still works other than that.

I'm going to order a bunch of flat insulators and bushings for just those 2 triacs. I can insert them without going through the trouble of removing 12 nuts and screws to remove the heat sink.

[TheDucks]

I bout my G3's as a kit (the board was assembled. The strain bar has the studs for the Greens. That strain bar does not touch the heat sink. (p10) No Green goes to the heat sink (BTW there are really 2. A right and a Left. there is a gap Again, only Blacks and Whites go to the Board section. My G2 (was bought used) is not handy for a look see.

[Orville]

22 minutes ago, TheDucks said:

I bout my G3's as a kit (the board was assembled. The strain bar has the studs for the Greens. That strain bar does not touch the heat sink. (p10) No Green goes to the heat sink (BTW there are really 2. A right and a Left. there is a gap Again, only Blacks and Whites go to the Board section. My G2 (was bought used) is not handy for a look see.

Yes, mine has the gap, a left and a right, both have screws going through them and the grounding bar, there is No insulators between the heatsinks or the ground bar. 

Most every device I have ever built that uses a device that has a tab at the top to mount to a heatsink is also a ground that is connected to the ground pin on[inside/internally to] the devices heatsink mounting tab.

I have always, unless specified differently in the device specs, used that heatsink mounting tab on top as a grounding point.

I just took one apart for repair, and the tabs on all 16 triacs and both heatsinks were grounded when tested with a continuity tester(a component of the multimeters I check voltage levels, resistance and capacitance values with).

I always double and triple check things I take apart, make notes, even take reference photo's, then facilitate the repair(s) needed.  Especially when I don't have a wiring diagram or schematic, and have to make sure I don't put it back together incorrectly.

Edited March 11 by Orville

[TheDucks]

I have no idea why mine are different. Mine came with an Assembly book (and an addendum for a newer strain bar) in addition to the G3 users guide.

TO-220 devices, the 'tab' is almost never ground. as a matter, the CTB controller only has the Neutral (white) cluster of FasTabs at the bottom and the HOT is what gets switched . I believe the TAB is the load (also middle pin) of the triac.

While I modified the location of the AC input clamp to allow use of the CAT5 dangles, I did not change any the AC terminals

https://drive.google.com/file/d/1qvsqcSCvZ4j05208BLlbFwV-JWT2Qrod/view?usp=drive_link

[Orville]

4 hours ago, TheDucks said:

I have no idea why mine are different. Mine came with an Assembly book (and an addendum for a newer strain bar) in addition to the G3 users guide.

TO-220 devices, the 'tab' is almost never ground. as a matter, the CTB controller only has the Neutral (white) cluster of FasTabs at the bottom and the HOT is what gets switched . I believe the TAB is the load (also middle pin) of the triac.

While I modified the location of the AC input clamp to allow use of the CAT5 dangles, I did not change any the AC terminals

https://drive.google.com/file/d/1qvsqcSCvZ4j05208BLlbFwV-JWT2Qrod/view?usp=drive_link

WARNING: TRIACS, or THEIR HEATSINKS SHOULD NOT BE CONNECTED TO GROUND!

I just went out and looked at the one CTB16PC I currently have in operation and it's a V1 and is set up just like yours in your photo.

But the one I took apart for repair was NOT an RTG LOR Unit and this one did not work.  I had purchased a couple of used controllers, 1 from a garage sale(person selling it had no idea what it was, or even used for), and I just found my notes on this one.  I bought 2 of these used and the one I took apart for repairs, that may be the reason it didn't work. It's one I got and set aside, just found it in a box a couple months or so ago, and it never worked, even though it said working when I purchased it(piece of tape on box with "works" on it), but it didn't. 

Didn't buy it online or from any forum user, it was found at a garage sale with some indoor odds and ends holiday decor, but nothing that would have been used with it

The ground and triac heatsinks had been connected to ground and from what I just looked up researching more info on what a Triac does and works, no part of it should have been grounded.  So those heatsinks were installed incorrectly.

Will NOT be putting it back together the way I found it, if at all, as I may have to trash this board, since it may be completely fried because of this.  I don't really need the unit, so will salvage the weatherproof box for some other equipment for holiday use, as well as the dangles and other useable components.

So my apologies to everyone, but I didn't build this unit, LOR didn't build it, no idea who did, it was bought used at a garage/estate sale, no manual and no idea of age or version at the time.

 

Remember:

TRIACS, or THEIR HEATSINKS SHOULD NOT BE CONNECTED TO GROUND!

[Steven]

It turns out I was wrong. The triacs did not fail with an internal short to to the tab. The tab is still insulated.

The problem was actually a mechanical problem caused by the triac leads being bent and extending past the edge of the circuit board. It probably happened much earlier and was not noticed because a test to see of one channel interfered with another was probably never done  Here's what the board looked like:

You can also see where these leads left marks on the heat sink:

After trimming the triac leads and reassembly, all channels are working correctly.

[Orville]

Glad you found and corrected the problem.  The smallest missed detail can cause some of the biggest headaches!🤕🤭😂

[k6ccc]

Good catch.  I've never paid any attention to that (but I'm not using many AC controllers).