Channels 1-8 not working

[Tim Fischer]

Hi all,

I just sent this to LOR kit support, but if anyone here has any ideas, I'm open!

I recently assembled a CTB16PCK kit. This is my first LOR controller that I've built "from scratch". Problem is, I'm having issues with the left bank (Channels 1-8). The right bank seems to work just fine.

I checked the following from the kit troubleshooting doc:
1) Left half of board is powered. Triacs show 120V between center pin and neutral.
2) Fuse is good (see above)
3) U4 is properly oriented, no pins folded under. I 'touched up' the solder joints here to no avail.
4) R15 & R16 seem fine.
5) Solder joints seem fine, although it's really hard to tell with such small holes.
6) U2 might be my problem... I installed it backwards the first time (never powered it up that way) and had to pull it and reinsert it -- some pins got bent. I straightened them and it seems to be fine in its socket, but maybe something isn't making proper contact?



What I'm really looking for is someone to say "measure this at resistor x" or "make sure blah blah blah measures as follows..." so I can trace what in the world I did wrong...

I have a multimeter and digital logic probe at my disposal...

Thanks!

-Tim

[cmoore60]

Tim,

Have you browsed the following troubleshooting guide?

This has helped me in the past.

http://www.lightorama.com/Documents/Kit_Troubleshooting_Guide.pdf

Chuck

Sorry just read the steps that seem to apply. Your answers seem to show you followed the above link.

[Tim Fischer]

Yeah, that's the troubleshooting doc I mentioned, but thanks for the help.

I'm open to any other ideas -- would like to get this this puppy working!

-Tim

[cmoore60]

Tim

Are you powering this board with 2 power input cords?

If not then you will need to put jumpers in. The jumper locations are marked on the board. Also, if only using one power input cord you only need one fuse.

Chuck

[Tim Fischer]

Jumpers are in to power both sides of the board from one cord (at least for now -- I may put in dual cords for the final enclosure).

There is 120V of power from the triac's center pins (on both sides of the board) to neutral.

Thanks, and keep the thoughts coming...

-Tim

[Tim Fischer]

Just heard back from Dan -- he's thinking that the issue is either with U2 (the micro) or U4. Evidently there are 8 shared data lines between U4 and U5 (controlling the 2 halves of the board) and those must be working since my 9-16 are working. There's an additional "data latch line' for each chip, and that's sounding like the suspect. Or I suppose it could be power/ground to U4 as well...

I'll be checking that out when I get a chance (I'm at work now...)

Not sure what pins this latch line is on, but hopefully I'll figure it out...

-Tim

[Dr. Jones]

if you have another pc board - Try swapping U2

This will narrow it down. I have already sent one board in to LOR and it had a bad U2 chip.

[Tim Fischer]

Dr. Jones wrote:

if you have another pc board - Try swapping U2

This will narrow it down. I have already sent one board in to LOR and it had a bad U2 chip.

I don't have another PC board, unfortunately. Dan gave me some more details on the pinouts and voltages to look for and I'll be trying that when I get a chance (have been gone all day, just got back home about 20 minutes ago...)

-Tim

[Tim Fischer]

OK, once again I proved I'm an idiot .

Evidently in my haste in assembling the kit, I didn't notice that channels 1-8 have their optos installed upside-down compared to channels 9-16. So I now have 8 channels with the opto's installed backwards...

So, a couple questions (I've already emailed Dan as well):

1) Did I smoke 8 triacs? Or should they be ok, provided I can switch them around? (Looking at the MOC3020 datasheet, I see both sides have a N/C pin so I'm hoping??)

2) Any tips on how best to desolder these pups? I don't think I've ever desoldered an IC and had it live to tell about it... I guess depending on the answer to #1, it may not matter...

This is why I should just buy stuff premade .

-Tim

[cmoore60]

Tim,

Try solder braid.

I would remover all the optos. Then reinstall one to see if all functions properly. If that works then do the others.

Chuck

[Dr. Jones]

I prefer a solder sucker

[Tim Fischer]

I have a desoldering iron and a solder sucker. I just ordered a spool of desoldering braid off of ebay as well. So I should have my options covered

The problem I've always had with DIPs is that it's hard to get all of the solder off of all the legs, and you can't heat up all the pins easilly to get the dip to slide out... But with 6 pins maybe I have a prayer of getting it to work (heaven forbid I'd have to desolder and reinsert a U4/U5 or U2 chip

-Tim

[Tim Fischer]

Well, I swapped one of the optos around -- it now works fine. Only 7 more to go.

Dan also emailed me afterward and confirmed they should have survived the mistake...

To make matters even more fun, I knocked my desoldering iron off the stand and it fell and burned a hole in my office carpet :dude:. This is going to be one expensive kit .

-Tim

[Denny]

Tim,

I also did this on one of my kits, but only one opto! After that, I put a little black mark on the opto as I kept confusing myself with the little sumbol vs. the dot. I use that compressed air in a can to desolder. Just watch where you do it, blows the solder quite well!

I don't think you smoked the triacs.

[Tim Fischer]

Well after a couple more missteps that I won't get into here (too embarassing ) I got the remaining 7 opto's desoldered last night. I only broke off one pin, and fortunately it was in a position that I could make the connection to the neighboring resistor that it's supposed to connect to with a small solder bridge. I ran out of time (worked until about 1am!-- but then I didn't start until about 11:45...) and will have to solder them back in and test another night (maybe tonight).

Things should work assuming I didn't fry them while desoldering...

It will be nice to have this kit finished, finally.

Lesson to all: When building the kits, FOLLOW THE INSTRUCTIONS TO THE LETTER AND DON'T MAKE ANY ASSUMPTIONS! It's taken me about as long to 'fix' the kit as it did to build it the first time!

-Tim

[Tim Fischer]

Finished it up tonight, up through installing the heavy-duty heat sinks. Works like a charm!

So this saga ends, I hope! Now onto building it into an enclosure...

Thanks all for the help, and especially thanks for Dan for the extremely prompt and detailed debugging assistance!

-Tim

[mmaness]

Well, we'll get to see how well my desoldering skills are. I built 4 PC kits this past weekend and last night I tested them. One didn't work and when I looked at it, I had the notch on the wrong side of both U4 and U5. Desoldering two 20 pin ICs isn't my idea of how to spend a weekend, but I think it can be done with the right tools and a whole lot of patience.

Although I have a desoldering vacuum thing, I think for this one I'm going by Radio Shack and getting their combo iron/desoldering bulb. It has the bulb attached to a 40w iron. That, plus braid and a bit of flux should (maybe) make this a successful chore.

If that doesn't work, then the only solution is to cut the legs off the chip, remove the legs from the board, and get two new chips. Personally, I'd rather give desoldering the old college try. I had success doing this quite a few years ago, but it does take some skill.

Next, I'm going to have to brush up on my reading skills. I successfully built 5 kits previously with no problems whatsoever, then all of a sudden do something as stupid as this.

AAARRRGGGHHHH!!!!

[Tim Fischer]

I feel your pain!

I'm almost embarassed to admit it, but I also had installed U2 (the micro) in backwards, but fortunately a) I noticed it before powering up the board, and it's in a socket!

Man, if only there were detailed instructions or something telling you which way these things went!

-Tim

[Denny]

Rather than desoldering, just mount the board upside down. That way, the notches will be pointing in the right direction!

Maybe it would be easier just to cut all the leads and then remove each lead individually. Then, get two new ICs from Dan; he will replace them free.

We need a "Murphy's Law" forum for our mistakes! When I got my first board, I ordered the assembled 1602, all I had to do was mount it in a box and attach the pigtails. Somehow, I managed to attach two hot leads to channel 9 and cooked the triac when I powered it up.

[Tim Fischer]

Denny wrote:

Maybe it would be easier just to cut all the leads and then remove each lead individually. Then, get two new ICs from Dan; he will replace them free.

I'm sure that would be easier -- but I felt bad intentionally ruining 8 optos and making Dan pay... I figured I'd ruin a few in the process and I'd only replace those, but as it ended up, I only broke one leg off one opto and I was able to make the connection with a solder bridge (it was the one between the high-voltage side of the triac and the resistor to AC-Hot...)

-Tim