What Did You Do Today???
- Thread starter Tony
- Start date
Tony
Staff member
But it's not empty though. I have tools off to the left of the screen plus my lamp. Then there is the picture frame, 3 action cameras under the center monitor, a switch, a soldering iron (that you all have not seen... yet) soldering paste, flux and flight controllers under the right monitor. Not to mention the hub on top of the switch lmao.
Tony
Staff member
Just fired up this old XP lappy again and I had a thought... I wonder how it will work if I find a way to install an SSD in it. I don't think they made an IDE SSD lmao. But may I would love to use this computer more often... I absolutely love this keyboard even though I can out type this poor thing....
Tony
Staff member
I finally got around to working on Matt's lights that he brought me to play with. He had a power surge at his house and both lights died. Thankfully, one of them was just the PSU which is just a 19 volt laptop power supply essentially. The other light had a still good power supply, but the light was dead. He told me about it and with me getting more and more into component level repair, I told him to bring them down. This was the same time he stole his T-Maxx back.... lmao
Well, I needed to order some things from China to do some testing so I had to wait quite a while before I could do any 'real' testing or even see some of the things on this board. Well that stuff arrived and it turns out, I needed NONE of it. Well, almost.
I had narrowed the issue down to a direct short to ground. I tried everything I knew to find this friggin short and I could not find it. I removed components and the short would go away, but the component would test good. Finally after a few hours of playing with this thing I said screw it. I had nothing more to lose, the light was toast regardless so it was time to force some voltage into it. One of the items I ordered to do this stuff was a regulated power supply (very cheap and not what I need, but it is all I have that I can regulate). It only goes to 15 volts and maxes out at 1 amp. So I connected my probes to the output and put the probes where the power wires went (I had removed them)
At first I started out at a half volt and watched the board for some magic smoke. Nothing.... I looked at the amps and even at a half volt it was pulling 1/4 of an amp. So I knew if I could get enough power pumped into it, the little sucker would either pop or smoke, so I kept going. I finally got up to 2 volts before my PSU started kicking off so I backed it off slightly until it would stay on and kept an amp going into it. But still, no smoke... So I took a cotton swab with alcohol on it (rubbing, not drinking) and started going around the board. The issue... It is covered in silicone. The light goes over a saltwater aquarium so it has to be coated. This causes the alcohol to just bead up and not show what is getting hot.
FINE!!!!!
I took my FINGER and started touching the board in different places to see if I could find something warm. Yup, that did it. I had been feeding voltage to this thing the entire time and when my finger hit the component, I knew it. Holy hell was that thing hot!!!! But guess what... I found my short to ground.
So what was it? It was an N-Channel Mosfet...
That little mosfet completely disabled a $400 light! As soon as I removed it, the light started working, but without the ability to control the color. It is stuck right in the middle so the same amount of white light as there is blue light. But I can adjust intensity and I think this is about where Matt had the color set. I guess I could hook it up to my apex and find out if I can adjust it through the controller...
I did order 3 of those little mosfets just in case I screw one up which is likely. Oh, and what was the cost of that little 3.4mm X 3.4mm X 1mm chip that killed a $400 light? $1.05..........
Well, I needed to order some things from China to do some testing so I had to wait quite a while before I could do any 'real' testing or even see some of the things on this board. Well that stuff arrived and it turns out, I needed NONE of it. Well, almost.
I had narrowed the issue down to a direct short to ground. I tried everything I knew to find this friggin short and I could not find it. I removed components and the short would go away, but the component would test good. Finally after a few hours of playing with this thing I said screw it. I had nothing more to lose, the light was toast regardless so it was time to force some voltage into it. One of the items I ordered to do this stuff was a regulated power supply (very cheap and not what I need, but it is all I have that I can regulate). It only goes to 15 volts and maxes out at 1 amp. So I connected my probes to the output and put the probes where the power wires went (I had removed them)
At first I started out at a half volt and watched the board for some magic smoke. Nothing.... I looked at the amps and even at a half volt it was pulling 1/4 of an amp. So I knew if I could get enough power pumped into it, the little sucker would either pop or smoke, so I kept going. I finally got up to 2 volts before my PSU started kicking off so I backed it off slightly until it would stay on and kept an amp going into it. But still, no smoke... So I took a cotton swab with alcohol on it (rubbing, not drinking) and started going around the board. The issue... It is covered in silicone. The light goes over a saltwater aquarium so it has to be coated. This causes the alcohol to just bead up and not show what is getting hot.
FINE!!!!!
I took my FINGER and started touching the board in different places to see if I could find something warm. Yup, that did it. I had been feeding voltage to this thing the entire time and when my finger hit the component, I knew it. Holy hell was that thing hot!!!! But guess what... I found my short to ground.
So what was it? It was an N-Channel Mosfet...
That little mosfet completely disabled a $400 light! As soon as I removed it, the light started working, but without the ability to control the color. It is stuck right in the middle so the same amount of white light as there is blue light. But I can adjust intensity and I think this is about where Matt had the color set. I guess I could hook it up to my apex and find out if I can adjust it through the controller...
I did order 3 of those little mosfets just in case I screw one up which is likely. Oh, and what was the cost of that little 3.4mm X 3.4mm X 1mm chip that killed a $400 light? $1.05..........
Tony
Staff member
So here is an update to the above post. I ordered 3 of those mosfets (just in case) from Digikey and wow shipping was smokin' fast! I received them yesterday and I promptly cleaned the pads, applied fresh leaded solder and some tacky flux and stuck the new mosfet on the board. I grabbed the hot air gun and heated it up to melt the solder, saw it move into place, swirled the hot air gun just to make it "dance" and seat better and that was it, the new mosfet was soldered on.
I plugged it in and BOOM, the light worked. Of course it worked without it as well. So the real test here is if I can control the color. I started turning the dial and NOPE, still stuck on one color which is a 50/50 mix of white and blue. Ugh....
So I hit up Matt (@callsign4223) and started asking him about wiring these lights to the apex so that I can see if I can adjust the color through the program. I know 0 voltage is all blue and 10 volts is all white with 5 volts being where it is stuck now. He proceeded to school me (as he usually does lmao) and told me to just plug a 2 ring jumper between the two lights and use the working light to control the light I just repaired as a slave.
Before I explain that, let me back up a little bit here. This board is weird and highly irritates me. As stated before the boards in this light have silicone covering them to protect them from saltwater. Well, when I was trying to narrow down the short to ground, I thought removing the output plug (the plug right there where the color adjustment knob is) would possibly remove the short. Not only is there clear silicone on these, but all sockets are covered in even stronger black silicone to secure them to the board. So I grabbed my solder wick and removed all of the solder I could and heated up each of the 5 legs of this plug to remove it.
Two issues here. First, I was pulling against the silicone that I could not cut with my exacto knife. This means I would have to pull just a little harder than I normally would. Second issue is, I could not see the solder under the socket which I had thought I removed. So yea, more force than required, not enough heat because the sockets were plastic and I didn't want to melt them, silicone pulling against me and you can guess what happened next.... Yup, pulled every friggin pad on this thing... But it's the output, who cares. Just use the input side and make this light the slave with the other light as the master and all is well.
Back on track...
I took the jumper off of my AI SOL Blue lights (just a 3.5mm 2 ring or 3 connection plug) and I plugged into the good light and the light I was working on. Good light = master, Light I'm working on = slave just as stated above. I started with trying to adjust intensity and that worked just fine! So I started playing with the color and the master was changing, but the slave was stuck at 50/50 still.... UGH.
I pulled out the multimeter and started checking the socket that I had removed (I didn't put it back on after installing the new mosfet, but removed it before finding the mosfet issue). What I noticed was there are only 3 of the legs that are used when it is plugged in, but unplugged two of those 3 are shorted to two of the other pins. I started probing around on the good light and these two legs that when a wire is plugged in do nothing, but shorted to the other two when unplugged, GO NOWHERE!!! There are absolutely no traces going to them. I'm 99% sure this is a 2 layer board because I probed around everywhere I could think of and I got no continuity between those two legs and anywhere else on the board. So this was a dead end right? Nope....
I decided to put the socket back in and run jumpers to repair the pads that I tore out. I hit every leg with a LOT of solder to secure it to the board and then I ran the jumpers where they needed to go. And for S&G's I plugged it into power and WTF, the color is now adjustable.... This shouldn't happen, but it did and I was happy. So I hit up Matt and told him to come get these lights, they are working great. While we were chatting and just as I was shooting a video of them the light I was working on goes out..... Nothing.... DEAD!!! UGH!!!
I pulled the power connector and checked voltage and there was none. This is what it would do before when there was a direct short. It would shut itself down to save the power supply. So I put the multimeter into continuity mode and checked the positive and negative on the power connector of the light and sure enough, dead short. So I grab my flux (tacky flux mind you, very thick) and was going to inject voltage into the light to see which one made the tacky flux go liquid and then I thought to just remove the mosfet that I had installed.
I grabbed the hot air and remove it, plugged the light back in and.... nothing... I forgot to reset the power supply lol. I unplugged the power supply from wall voltage (mains for those across the pond) for a while and plugged it back in and BOOM, the light works... I turned the color knob and yup, I can still adjust color. As far as I can see that mosfet does nothing, but can completely kill the light and something on the board is killing it. But it works fine without it.... I'm not sure what is going on there, but hey, the light is working and that is all I care about.
Now, Matt can put these very expensive lights back over his tank and he can make his tank great again lmao. I have a feeling he will be hitting me up for some more frags now that he has lights again.
I do have to say, this was a fun project and I'm glad he brought them down to me to play with.
I plugged it in and BOOM, the light worked. Of course it worked without it as well. So the real test here is if I can control the color. I started turning the dial and NOPE, still stuck on one color which is a 50/50 mix of white and blue. Ugh....
So I hit up Matt (@callsign4223) and started asking him about wiring these lights to the apex so that I can see if I can adjust the color through the program. I know 0 voltage is all blue and 10 volts is all white with 5 volts being where it is stuck now. He proceeded to school me (as he usually does lmao) and told me to just plug a 2 ring jumper between the two lights and use the working light to control the light I just repaired as a slave.
Before I explain that, let me back up a little bit here. This board is weird and highly irritates me. As stated before the boards in this light have silicone covering them to protect them from saltwater. Well, when I was trying to narrow down the short to ground, I thought removing the output plug (the plug right there where the color adjustment knob is) would possibly remove the short. Not only is there clear silicone on these, but all sockets are covered in even stronger black silicone to secure them to the board. So I grabbed my solder wick and removed all of the solder I could and heated up each of the 5 legs of this plug to remove it.
Two issues here. First, I was pulling against the silicone that I could not cut with my exacto knife. This means I would have to pull just a little harder than I normally would. Second issue is, I could not see the solder under the socket which I had thought I removed. So yea, more force than required, not enough heat because the sockets were plastic and I didn't want to melt them, silicone pulling against me and you can guess what happened next.... Yup, pulled every friggin pad on this thing... But it's the output, who cares. Just use the input side and make this light the slave with the other light as the master and all is well.
Back on track...
I took the jumper off of my AI SOL Blue lights (just a 3.5mm 2 ring or 3 connection plug) and I plugged into the good light and the light I was working on. Good light = master, Light I'm working on = slave just as stated above. I started with trying to adjust intensity and that worked just fine! So I started playing with the color and the master was changing, but the slave was stuck at 50/50 still.... UGH.
I pulled out the multimeter and started checking the socket that I had removed (I didn't put it back on after installing the new mosfet, but removed it before finding the mosfet issue). What I noticed was there are only 3 of the legs that are used when it is plugged in, but unplugged two of those 3 are shorted to two of the other pins. I started probing around on the good light and these two legs that when a wire is plugged in do nothing, but shorted to the other two when unplugged, GO NOWHERE!!! There are absolutely no traces going to them. I'm 99% sure this is a 2 layer board because I probed around everywhere I could think of and I got no continuity between those two legs and anywhere else on the board. So this was a dead end right? Nope....
I decided to put the socket back in and run jumpers to repair the pads that I tore out. I hit every leg with a LOT of solder to secure it to the board and then I ran the jumpers where they needed to go. And for S&G's I plugged it into power and WTF, the color is now adjustable.... This shouldn't happen, but it did and I was happy. So I hit up Matt and told him to come get these lights, they are working great. While we were chatting and just as I was shooting a video of them the light I was working on goes out..... Nothing.... DEAD!!! UGH!!!
I pulled the power connector and checked voltage and there was none. This is what it would do before when there was a direct short. It would shut itself down to save the power supply. So I put the multimeter into continuity mode and checked the positive and negative on the power connector of the light and sure enough, dead short. So I grab my flux (tacky flux mind you, very thick) and was going to inject voltage into the light to see which one made the tacky flux go liquid and then I thought to just remove the mosfet that I had installed.
I grabbed the hot air and remove it, plugged the light back in and.... nothing... I forgot to reset the power supply lol. I unplugged the power supply from wall voltage (mains for those across the pond) for a while and plugged it back in and BOOM, the light works... I turned the color knob and yup, I can still adjust color. As far as I can see that mosfet does nothing, but can completely kill the light and something on the board is killing it. But it works fine without it.... I'm not sure what is going on there, but hey, the light is working and that is all I care about.
Now, Matt can put these very expensive lights back over his tank and he can make his tank great again lmao. I have a feeling he will be hitting me up for some more frags now that he has lights again.
I do have to say, this was a fun project and I'm glad he brought them down to me to play with.
Tony
Staff member
It actually wasn't all that bad. At first I was just trying to probe and find the short, but that was doing nothing. And as we all know, if you force voltage into something that is direct shorted, something is going to get hot (look at your toaster, oven, range and so on). So that is what I had to do. The frustrating part of it is that it destroyed a brand new chip within 5 minutes of being on at its lowest setting and I don't know what is causing that. And as long as they work like normal, I never will lol. Might put it over my tank and let it run with my lights just to test before he picks it up next weekend possibly.
Tony
Staff member
I need to check if it is the gate or the source that is shorting to ground (drain) and which side it is on since this one has two sides. I REALLY wish Kessil wasn't such asshats and just cough up the schematics so I can see where these things are going. But, they won't, I tried...