Apollo 6 / Photon 90 - A look inside and looking for alternatives ways of cooling

Rasser said:

Looks perfect to me, if you want to control the 3 drivers individual that is, else I would look into incorporating the 4 blue LED into the 2 other stings.

Click to expand...

That's why I was confused because they have different voltages I didn't know if I could add three lower volt than one high volt then 3 lower volt all on one driver just because the current is the same they both have different voltage range

So a 2.6v -2.6v - 4.2v - 2.6v -2.6v
i would be runnin 700ma through all of them but woul the voltage go up to the 4.2 and drop back down to the 2.6 if I ran them in series Cus from what I though I couldn't do that

DO3SHA said:

So a 2.6v -2.6v - 4.2v - 2.6v -2.6v
i would be runnin 700ma through all of them but woul the voltage go up to the 4.2 and drop back down to the 2.6 if I ran them in series Cus from what I though I couldn't do that

Click to expand...

When connecting LED's with different voltage in series, it's a bit like connecting resistors in series with different value,
the more current that flows through, the larger the voltage drop over them becomes.

I recommend this Circuit Simulator : It don't look at much but has a tons of build in circuits templates and the possibilities are endless and has current animations.

I've made this example of two blue LED 3,5V and two red LED 2,4V:


Note the yellow line is the current and the green is voltage.

If you go to the menu and select import and put this text in the field, the LED simulation circuit in the image should load, I think.

$ 1 5.0E-6 10.20027730826997 63 10.0 62
v 144 384 144 64 0 0 40.0 12.0 0.0 0.0 0.5
w 144 64 416 64 0
w 144 384 272 384 0
O 464 64 528 64 1
w 416 64 464 64 0
w 272 384 464 384 0
O 464 144 528 144 1
O 464 224 528 224 1
O 464 304 528 304 1
162 464 304 464 384 1 2.4 1.0 0.0 0.0
162 464 224 464 304 1 2.4 1.0 0.0 0.0
162 464 144 464 224 1 3.5 0.0 0.0 1.0
162 464 64 464 144 1 3.5 0.0 0.0 1.0
o 12 64 0 35 5.0 3.2 0 -1
o 11 64 0 35 5.0 3.2 1 -1
o 10 64 0 35 5.0 3.2 2 -1
o 9 64 0 35 5.0 3.2 3 -1

Click to expand...

Right click for all the editing stuff and more.

Rasser said:

When connecting LED's with different voltage in series, it's a bit like connecting resistors in series with different value,
the more current that flows through, the larger the voltage drop over them becomes.

I recommend this Circuit Simulator : It don't look at much but has a tons of build in circuits templates and the possibilities are endless and has current animations.

I've made this example of two blue LED 3,5V and two red LED 2,4V:

Note the yellow line is the current and the green is voltage.

If you go to the menu and select import and put this text in the field, the LED simulation circuit in the image should load, I think.



Right click for all the editing stuff and more.

Click to expand...

awsome man thanks, I was jsut confusing myself when i knew it all along, i guess thats what happens when i dont smoke and do this.

is anyone gettign a restrict access when posting and have to fill out the number letter box to verify?

DO3SHA said:

awsome man thanks, I was jsut confusing myself when i knew it all along, i guess thats what happens when i dont smoke and do this.

Click to expand...

Thanks your self, I did not know the program could exchange circuit like this, and has all these functions.

In order to simulate a LED right the program has to know it's IV curve(The charts I've been making)
so it properly just have one default. so changing the voltage value is really not enough but for teaching and simulating it's fine.

DOH3SHA > "is anyone gettign a restrict access when posting and have to fill out the number letter box to verify? "

Yes, I have, pain in the but reading these letters.

Rasser said:

DOH3SHA > "is anyone gettign a restrict access when posting and have to fill out the number letter box to verify? "

Yes, I have, pain in the but reading these letters.

Click to expand...

its been a pain in the but having to do it form the phone, damn apple wont allow the rollitup app

The circuit I've designed is drawing way to much current 1.7A
so if this had not been a simulation then the 3W LED's would be toast.

If someone wants a challenge we could say that the 12 volt are fixed,
so we have to fit a resistor in series with the right value,
so the current is close to say 700mA and see how big a wattage
that resistor has to manage, witch say something about energy wasted and cost of the resistor.

nb. You edit nodes/wires in the sim by holding 'ctrl' while moving the mouse.
Press space after placing a component to move a selected object in stead of placing more components.


I've already done it here, for those not up to a challenge, but still want to see it run.

$ 1 5.0E-6 10.20027730826997 63 10.0 62
v 144 384 144 64 0 0 40.0 12.0 0.0 0.0 0.5
w 144 64 320 64 0
w 144 384 272 384 0
O 464 64 528 64 1
w 384 64 464 64 0
w 272 384 464 384 0
O 464 144 528 144 1
O 464 224 528 224 1
O 464 304 528 304 1
162 464 304 464 384 1 2.4 1.0 0.0 0.0
162 464 224 464 304 1 2.4 1.0 0.0 0.0
162 464 144 464 224 1 3.5 0.0 0.0 1.0
162 464 64 464 144 1 3.5 0.0 0.0 1.0
r 320 64 384 64 0 0.47
o 12 64 0 35 5.0 0.8 0 -1
o 11 64 0 35 5.0 0.8 1 -1
o 10 64 0 35 2.5 0.8 2 -1
o 13 64 0 35 1.25 0.8 3 -1

Click to expand...

Rasser said:

To get a better overview and to illustrate the valid voltage combinations that's possible when using more then on DC source in series.
And as an example the use of a 10W LED constant current driver and PC PSU's to driver more then 1 LED with the same driver.
But note the PC PSU use 14 watt at idle so it only make sense if there is a heavy draw 100 watt per PSU or so.
All LED drivers lose 5-15W per 100W, the ones in my 10W LED flood light looks like those that has a 85% efficiency, so using many of them
also has a price. ~1W per driver.

The color codes should points to the right reference GND to use with the right PSU and voltage.
Since we are interested in the highest voltage it's the +12 & GND thats connected between the PSU's

In the example used they both run on +36V one of them could use +29V by being connected to +5V on PSU 3, if that voltage was required.

Besides the "main" power source(PSU3) using this as a external power station there is a lot of voltage possibilities to take from:
+5V(PSU1,2,3) to the temperature displays and such(USB). +12V(PSU1,2,3) to the fan controller. +12V(PSU1,2,3) to DC to DC constant current drivers,
like the ones used in MR16 12V LED spots.

Click to expand...

I'm happy to see that my new EPS 550W PSU's are not drawing 14 watts like the one a bit older, but only 8W,
so I've updated the image above, for it change the situation a bit now it would make sense to only draw 70-100W from a single PSU vs. 160-200W
when considering the energy efficiency of the unit.

And when measuring the PSU's +12V line I was reading 12.2V without a load, and 12.0V when running my four 10W chips and it stayed there all the time.
To bad the chips only draw 370mA at 6V otherwise a single PSU could run a shit load of them.

So I've looked a bit into making the PC PSU's adjustable or lifting the voltage, this lousy video makes it real tempting, if I could get 16.5V from a single PSU, two of them could run all my Apollo modules.

Mod your ATX power supply into an adjustable lab power supply!

[video=youtube;mK_rIs7uFlo]http://www.youtube.com/watch?v=mK_rIs7uFlo[/video]


I tried to mod mine but so far without luck, fricking SMD components are so tiny and impossible to work with,
I changed a smd resistor to a potentiometer on wires, but turning that don't get the voltage to move, and it just shuts down at a point,
but I haven't giving up on that subject yet.

Thanks, interesting when people take apart and show us