DIY with Quantum Boards

Chew420

Chew420

Member
What is the differnce in the HLG-240H-C2100B and the ELG-240-C2100B for use for 2 hlg 288s. Also the ELG-120-54 and the HLG-120H-54B for use with QB 96? I need drivers for these lights and i do not know much about difference in each driver. I have potentiometers for dimming to hook up, what do you Good guys and Good girls who have knowledge recommend? dimming is important to me as well as brightness maybe a brighter driver for the qb 96. Im pretty new here help would be much appreciated thanks.
 
Moabfighter

Moabfighter

Well-Known Member
daveybc said:
In my 4x8 tent they are mounted the 4' span as seen here.

Since there was an angled roof and limited hanging height we figured not to go longer than 22" max on the design. My buddy was replacing the ViperSpectra600 he had. We installed 3 of those 260 fixtures so far, my buddy is blown away and the driver was set to about 2/3 power. Once all is done, ill be sure to put up a pic.
Click to expand...
Hella qb96 wow. Nice.
 
pop22

pop22

Well-Known Member
I'd be curious to see, how much if any difference is made by the thermal paste vs without. On CPU heatsinks it can make a difference of several degrees, but in that application your trying to dissipate heat from a small, radiating surface onto a larger surface. It likely will make a difference, but not sure it will be enough to make much difference.

Federucci said:
It depends on the wire and how long you run it. I have a run of about 15 feet (est). I am using a 14/3 power cord to input and 14/3 on output until just before the unit where I switch to NTE 18 gauge solid. It's not necessary, but it will save about .2-.25 volts. I will run the green conductor of the 14/3 as a ground with an M4x6mm button head where the bolt hole is on the slate 2 where you typically mount the driver to is. I might as well do it, even if it's not necessary.

I used this calculator to help make my decision on what to run: https://www.calculator.net/voltage-drop-calculator.html?material=copper&wiresize=8.286&voltage=120&phase=dc&noofconductor=1&distance=10&distanceunit=feet&amperes=2&x=59&y=24

Speaking of unnecessary, I will be trying "high end" thermal paste on the back of the QB648 boards as each bank of 2 will be connected to a 320H-C2800AB driver, even though I will probably be running them at around 125w max with a 140mm fan on each individual board. I wonder if I can take these boards to the driver limit of about 175w/board as they are rated 56v at 6amps.

"Looks like she put a lotta mustard on that one!"
Click to expand...
 
F

Federucci

Well-Known Member
pop22 said:
I'd be curious to see, how much if any difference is made by the thermal paste vs without. On CPU heatsinks it can make a difference of several degrees, but in that application your trying to dissipate heat from a small, radiating surface onto a larger surface. It likely will make a difference, but not sure it will be enough to make much difference.
Click to expand...
I think it depends on the quality of TIM and how you apply it. Steve at HLG advised that running 2xQB648s on a 320H-C2800 driver would be "running them pretty hard" and to think about overall thermal management.

Damned nerds and their newfangled "overclocking conventions".
 

Attachments

Last edited:
F

Federucci

Well-Known Member
Higher Authority said:
Single 648 at 2.8a on slate 2 with no TIM.

View attachment 4558191


That would be a lot of paste...but I can try it and see the difference. I would not use liquid metal on aluminum PCB's. I only have a little Kryonaut left, not enough for a 648. Will get some cheap artic silver and see the difference and move to better paste for the sake of testing if AS5 shows valuable gains.
Click to expand...
Any active cooling?
 
HolyAngel

HolyAngel

Well-Known Member
Higher Authority said:
Single 648 at 2.8a on slate 2 with no TIM.

View attachment 4558191


That would be a lot of paste...but I can try it and see the difference. I would not use liquid metal on aluminum PCB's. I only have a little Kryonaut left, not enough for a 648. Will get some cheap artic silver and see the difference and move to better paste for the sake of testing if AS5 shows valuable gains.
Click to expand...
Oh nice flir pic! Yeah it takes a good ~10g of thermal paste to cover a board. I did it for both of my 288 rspecs on a 320h and took a little over 1 20g tube of paste to do both.
 
F

Federucci

Well-Known Member
HolyAngel said:
Oh nice flir pic! Yeah it takes a good ~10g of thermal paste to cover a board. I did it for both of my 288 rspecs on a 320h and took a little over 1 20g tube of paste to do both.
Click to expand...
It all depends if paying another $25/board for 10g or so of the high end stuff juice is worth the squeeze. I think that since we're all running, or going to be running the 320h with 2 QBs, and it has been known to draw 10% over the designated wattage, then it would be prudent to use a good quality thermal paste along with a 120/140mm fan blowing on top of each board. I cant see any reason not to. From my best guess and research, the average thermal conductivity of anodized aluminum is somewhere around 20W/m-K, but I am not sure that the TC of the Slate boards actually is. I think it has to be somewhere around 20. I don't even know what this means in terms of real world performance yet. I'm sure that raw aluminum is more thermally conductive, but

One thing I learned is that the conductivity "rating" does not necessarily translate well into real-world results. Some companies like Noctua do not list their W/mk ratings but tend to perform a little better than pastes which also come highly recommended, like MX-4. Viscosity is also a big thing. You can get a thinner layer with something like Kryonaut than with PK-3, so it would stand to reason that you could possibly get away with a bit less than the 10g with the Kryonaut than say Noctua or MX-4.

Time to sign up for an "overclocking convention" and bring my shitty 2013 Lenovo, show these amateurs how a REAL MAN overclocks a low end laptop than can't even load youtube well.
 
HolyAngel

HolyAngel

Well-Known Member
Federucci said:
It all depends if paying another $25/board for 10g or so of the high end stuff juice is worth the squeeze. I think that since we're all running, or going to be running the 320h with 2 QBs, and it has been known to draw 10% over the designated wattage, then it would be prudent to use a good quality thermal paste along with a 120/140mm fan blowing on top of each board. I cant see any reason not to. From my best guess and research, the average thermal conductivity of anodized aluminum is somewhere around 20W/m-K, but I am not sure that the TC of the Slate boards actually is. I think it has to be somewhere around 20. I don't even know what this means in terms of real world performance yet. I'm sure that raw aluminum is more thermally conductive, but

One thing I learned is that the conductivity "rating" does not necessarily translate well into real-world results. Some companies like Noctua do not list their W/mk ratings but tend to perform a little better than pastes which also come highly recommended, like MX-4. Viscosity is also a big thing. You can get a thinner layer with something like Kryonaut than with PK-3, so it would stand to reason that you could possibly get away with a bit less than the 10g with the Kryonaut than say Noctua or MX-4.

Time to sign up for an "overclocking convention" and bring my shitty 2013 Lenovo, show these amateurs how a REAL MAN overclocks a low end laptop than can't even load youtube well.
Click to expand...
This is exactly my setup. I did Gelid Extreme paste but Arctic MX-4 works too, and have a 140mm fan sitting a few inches above each heatsink. I don't think the boards breaks 55c at 374w from the wall but i'd have to double check.
 
F

Federucci

Well-Known Member
HolyAngel said:
This is exactly my setup. I did Gelid Extreme paste but Arctic MX-4 works too, and have a 140mm fan sitting a few inches above each heatsink. I don't think the boards breaks 55c at 374w from the wall but i'd have to double check.
Click to expand...
Most excellent. That's a lot of mustard. What did you use as a spacer for your fans? I have been thinking about what's the best way to mount them so they are more effective over a larger area.

I'm just wondering if you could get away with a thinner layer of a lower viscosity thermal paste so you can stretch out the value. Maybe use 7-8 grams per board instead of 10-11. At about $25 for 11.1 grams of Thermal Grizzly Kryonaut, I'm wondering if I can get away with 55.5g worth of those for 6 boards total. This stuff isn't that cheap, but it's supposedly the best. It's less viscous than the average TIM paste, so I am wondering if its performance is due to a much thinner layer contributing to the conductivity and acting as less of an "insulator".

If you all ever wanted to learn way too much about TIMS to get ready for that big overclocking convention, this is the place to do it:
www.tomshardware.com

Best Thermal Paste for CPUs 2024: Over 50 Compounds Tested and Ranked

The right thermal paste will improve performance.
www.tomshardware.com www.tomshardware.com
 
LarsVegasNirvana

LarsVegasNirvana

Well-Known Member
I don't think thermal paste is going to help. The limit is the heatsink's surface area, not thermal conductivity. You'd be better off spending your money on fans. Then again, I am using the 240w driver because it's better to just buy twice as many boards and heatsinks and run them softer.
 
HolyAngel

HolyAngel

Well-Known Member
Federucci said:
Most excellent. That's a lot of mustard. What did you use as a spacer for your fans? I have been thinking about what's the best way to mount them so they are more effective over a larger area.

I'm just wondering if you could get away with a thinner layer of a lower viscosity thermal paste so you can stretch out the value. Maybe use 7-8 grams per board instead of 10-11. At about $25 for 11.1 grams of Thermal Grizzly Kryonaut, I'm wondering if I can get away with 55.5g worth of those for 6 boards total. This stuff isn't that cheap, but it's supposedly the best. It's less viscous than the average TIM paste, so I am wondering if its performance is due to a much thinner layer contributing to the conductivity and acting as less of an "insulator".

If you all ever wanted to learn way too much about TIMS to get ready for that big overclocking convention, this is the place to do it:
www.tomshardware.com

Best Thermal Paste for CPUs 2024: Over 50 Compounds Tested and Ranked

The right thermal paste will improve performance.
www.tomshardware.com www.tomshardware.com
Click to expand...
I actually used some velcro and tied them to two rungs of the hangers. Here's a crappy pic
c545bbdf56633f13313860f482c192392a97f8c2.jpg

@LarsVegasNirvana
The thermal paste absolutely makes a huge difference, I saw well over a 10C drop in temps by having thermal paste and that was at half wattage.
 
R

RamenToppings

Member
Barristan Whitebeard said:
Ok, so you'll be wiring two boards to each driver in parallel. Do you want to mount your drivers on the heatsinks or remote mount them?
Do you have (preferably) two colors of 18 gauge solid core wire? Are you using wago connectors? The 3 port Wagos will work well.

I'll use red(positive) and black(negative) wire for an example. Driver V+ into 1st port of wago #1. 2nd port of wago #1 red wire to board #1 positive molex connector. 3rd port of wago #1 red wire to the board #2 positive connector. Now Driver V- into 1st port of wago #2. 2nd port of wago #2 black wire to board #1 negative connector. 3rd port of wago #2 black wire to board #2 negative connector. That's the DC side, just repeat these steps with the second driver and pair of boards.

If you remote mount your drivers you'll need to extend your wires on the DC side and depending on the distance between driver and lights, you may want to use a thicker gauge of wire for that run.

Did you buy potentiometers or are you going to use the internal potentiometer to adjust current?

For the AC side are you putting a plug on the driver input or are you using a power cord and a waterproof connector? Do you know which colors connect to which on the AC side or do you need help with that?
Click to expand...
1. Do you want to mount your drivers on the heatsinks or remote mount them?:
I was going to put the drivers on top of the heatsinks, is this safe?

2. do you have two colors of 18 guage solid core wire?
Yes

3. Are you using wago connectors?:
Yes

4. Did you buy potentiometers or are you going to use the internal potentiometer to adjust current?:
I did not buy potentiometers. I am not sure what this means or how to adjust the internal current.

5. For the AC side are you putting a plug on the driver input or are you using a power cord and a waterproof connector? Do you know which colors connect to which on the AC side or do you need help with that?:
I do not know and would need help with that.

EDIT: I am not sure what you mean by AC side
 
F

Federucci

Well-Known Member
Is 16 gauge solid copper too large to fit the Molex connectors on the QB boards? I know it says 18-22, but these jabronis are taking forever(4 weeks) and wanted to hear from you guys since that's what I have around.
 
pop22

pop22

Well-Known Member
I ran my own computer shop and was one of those nerds lol! Gave all that up after I retired. I built a few nice, watercooled, overclocked machines. We'd benchmark them then sell them. Ran a cpu with no heatsink.......... 2 seconds and it went POOOOF! hehehe

Federucci said:
I think it depends on the quality of TIM and how you apply it. Steve at HLG advised that running 2xQB648s on a 320H-C2800 driver would be "running them pretty hard" and to think about overall thermal management.

Damned nerds and their newfangled "overclocking conventions".
Click to expand...
 
You must log in or register to reply here.
Top