ClobberWatts
Well-Known Member
This is going to take a bit to write out, but I think it should be done for posterity. Small checklist of things I am going through to get this thing running right.
First off. This is for vero 29 se's with 140mm heatsinks. Don't know if it applies to cree's, citizen's, boards, or other chips, but I'm sure some of it will apply.
Here's the setup before ripping it apart.
I should not have put the refelctor holders on before wiring it together and testing it. I thought since I knew where the positives are and negatives are, that I wouldn't run into any problems with the wiring or poke in connections. That's true and not at the same time. I wanted a good solid contact between the chip, thermal paste, and heatsink. Problem with the Vero's is that the reflector holder uses the same screw as the chip covering the poke in disconnects. Looking back, my wanting of a solid thermal paste is not as important as making sure that every connection is solid and a full signal is going through the whole system first, if that makes sense. The paste should glue it enough for you to test it first with a quick fire up, then unscrew your chips, add the holders, and put it together finally. If your using pads, this probably isn't so much of a problem.
Alright, so I tested the system before I ripped it apart and did not get a signal through the eight chips or when graytail told me to drop down to seven. Problem was with that is.. I don't remember if I tried it backwards which I will explain here in a sec. So I took the whole thing apart down to just the chips and every wire seperate. All wires gave me 0 ohms. Wago's tested good and tried different one's just in case. All chips at first gave me open line or OL. But when I put the positive of my meter on the negative of the chip and the negative/common of my meter on the positive of the chip, this is the resistance I got on the chips from the photo above from top left.
1.1 .7
1.05 .5
.95 .5
.5 .5
Tested connections with just my meter and realized that 1 was now 0. Now I have no idea what those numbers mean, but it does make me think for another reason why I would never run chips in parallel. Just opinion here. But these chips are too sensitive and each one is so different. You couldn't get the exact voltage to run exactly for each chip in parallel if I'm understanding it correctly. It also told me at least that I'm getting some sort of signal through the chips and each one was so different and fluctuating which made me come up with some other weird theory... like the five's were getting more sunlight from the window than the others and if you look at the angle. Top left was the highest, the furthest from the angle of sunlight. Were these things acting like solar panels. Look at the angle of measurements. Weird.
All right, back to the task at hand. Now that I know it's not my chips or one chip for that matter. Not the wiring or the poke in connection since once I put it back together and doing my reverse ohm test. I got zero through the system. Then I thought. Ok. My potentiometer is testing @ 101k, but what if it's not sending right signal. Think it was Fanleaf that said if you disconnect one wire it will pull full power which didn't make sense to me. It would seem that you would have to connect both wires together on the dimmer lead to get full power. Didn't matter. Tried it both ways and still no light.
At this point. I'm getting 333v on the dc line from the driver still as I said in previous post. I know I have solid connection on all wires and chips, so I connected a different extenstion cord straight to the driver without the potentiometer or my box with fuses. Still... no light.
My next step that I am working on now is to hook up the 240h-c2100b to three chips on each side and then the final two. If it fires up, I gotta put plugs and shit on it tonight/late first, then I'll know it's a bad 480h, but that still wouldn't explain why I would be getting the 333v on the DC.
We shall see said the blind man.
First off. This is for vero 29 se's with 140mm heatsinks. Don't know if it applies to cree's, citizen's, boards, or other chips, but I'm sure some of it will apply.
Here's the setup before ripping it apart.
I should not have put the refelctor holders on before wiring it together and testing it. I thought since I knew where the positives are and negatives are, that I wouldn't run into any problems with the wiring or poke in connections. That's true and not at the same time. I wanted a good solid contact between the chip, thermal paste, and heatsink. Problem with the Vero's is that the reflector holder uses the same screw as the chip covering the poke in disconnects. Looking back, my wanting of a solid thermal paste is not as important as making sure that every connection is solid and a full signal is going through the whole system first, if that makes sense. The paste should glue it enough for you to test it first with a quick fire up, then unscrew your chips, add the holders, and put it together finally. If your using pads, this probably isn't so much of a problem.
Alright, so I tested the system before I ripped it apart and did not get a signal through the eight chips or when graytail told me to drop down to seven. Problem was with that is.. I don't remember if I tried it backwards which I will explain here in a sec. So I took the whole thing apart down to just the chips and every wire seperate. All wires gave me 0 ohms. Wago's tested good and tried different one's just in case. All chips at first gave me open line or OL. But when I put the positive of my meter on the negative of the chip and the negative/common of my meter on the positive of the chip, this is the resistance I got on the chips from the photo above from top left.
1.1 .7
1.05 .5
.95 .5
.5 .5
Tested connections with just my meter and realized that 1 was now 0. Now I have no idea what those numbers mean, but it does make me think for another reason why I would never run chips in parallel. Just opinion here. But these chips are too sensitive and each one is so different. You couldn't get the exact voltage to run exactly for each chip in parallel if I'm understanding it correctly. It also told me at least that I'm getting some sort of signal through the chips and each one was so different and fluctuating which made me come up with some other weird theory... like the five's were getting more sunlight from the window than the others and if you look at the angle. Top left was the highest, the furthest from the angle of sunlight. Were these things acting like solar panels. Look at the angle of measurements. Weird.
All right, back to the task at hand. Now that I know it's not my chips or one chip for that matter. Not the wiring or the poke in connection since once I put it back together and doing my reverse ohm test. I got zero through the system. Then I thought. Ok. My potentiometer is testing @ 101k, but what if it's not sending right signal. Think it was Fanleaf that said if you disconnect one wire it will pull full power which didn't make sense to me. It would seem that you would have to connect both wires together on the dimmer lead to get full power. Didn't matter. Tried it both ways and still no light.
At this point. I'm getting 333v on the dc line from the driver still as I said in previous post. I know I have solid connection on all wires and chips, so I connected a different extenstion cord straight to the driver without the potentiometer or my box with fuses. Still... no light.
My next step that I am working on now is to hook up the 240h-c2100b to three chips on each side and then the final two. If it fires up, I gotta put plugs and shit on it tonight/late first, then I'll know it's a bad 480h, but that still wouldn't explain why I would be getting the 333v on the DC.
We shall see said the blind man.
great job !!!