V series "Tetras"

[robincnn]

If it stays OFF or On ,the fan has failed and no longer operates.
So .. A blue small led ,for the "Fan OK " thing and a bi-colored ,red-green ,for the "SoftOnSet " thingy ...
At the center ,the 12 position dial ,of the single pole rotary switch adjusting
the driver's output current .

Thanks for your inspiring designs.

Was there any reason you chose HLP-60H-42 over lpf 60d 42. Other than efficiency

Most quality fans have MTBF over 150k hours. Is fan relibilty still an issue?

I was thinking of a diy that will use bi metal thermal switch and will cut down power to 20% if heatsink temp goes over a threshold(fan failure).
Was thinking of adding 25k ohm and thermal switch parallel to 100k pot. Problem is I need thermal switch needs to stay open and close at the threshold temp, opposite of what those big metal thermal switches. Not sure if there is an easier way to do this.

 

[guod]

Normally Open (NO) is the secret here...
Normally Open Thermostat Switch - 50 Degrees
http://www.jaycar.com.au/Passive-Components/Circuit-Protection/Thermal-Switches-&-Fuses/Normally-Open-Thermostat-Switch---50-Degrees/p/ST3831

 

[Doer]

So you've become bored with making COB fixtures and are now working on building time machines.. how did I not see this coming?

It does look HG Wells

 

[salmonetin]

...thanks for the post Guod...

...on contactors way... ...industrial control boxes... for recyclate some parts...

...Contacts Normally Open (NO) ...normally used for conect...
...Contacts Normally Closed (NC) ...normally used for disconect...

saludos

 

[stardustsailor]

Thanks for your inspiring designs.

Was there any reason you chose HLP-60H-42 over lpf 60d 42. Other than efficiency

Yes..It happens that I tend to like open-frame drivers...
For example :
They seem to run-much- cooler ,than their encased counterparts ....
And...
Another example of reasoning ...
Easier to inspect and replace any blown -up electrolytic capacitor* ,ensuring drivers -prolonged service life .

( * Electrolytic Capacitors : #1 reason of a led driver failing .
Most sensitive parts and prone to
reduction of their service life,as their temperature increases.

)

Most quality fans have MTBF over 150k hours. Is fan relibilty still an issue?

Mean time between failures (MTBF) :
MTBF value prediction is an important element in the development of products.
However, it is incorrect to extrapolate MTBF to give an estimate of the life time of a component,
which will typically be much less than suggested by the original MTBF due to the much higher failure rates in the "end-of-life wearout" part of the "bathtub curve".

http://en.wikipedia.org/wiki/Mean_time_between_failures

I was thinking of a diy that will use bi metal thermal switch and will cut down power to 20% if heatsink temp goes over a threshold(fan failure).
Was thinking of adding 25k ohm and thermal switch parallel to 100k pot. Problem is I need thermal switch needs to stay open and close at the threshold temp, opposite of what those big metal thermal switches. Not sure if there is an easier way to do this.

Guod,has already suggested an ideal and simple solution .


Cheers.

 

[stardustsailor]

And some more about the "FAN OK " blinking LED feature ...

The circuit schematics :


How it works :
The incoming TACH signal is pulled up with a 4.7 K resistor.
Then it inputs at the CD4017 decade counter ,at pin 14 (CLK ) .
As fan is operating ,is producing two pulses per single full rotation .
A 1200 rpm fan ,then will be signaling 2400 pulses per minute .
that is 2400 / 60 =40 pulses per second .
Because only the output 9-of the decade counter- ,via a switch transistor is operating the LED ,
then the LED will only light once every 10 pulses.
The counter here acts as a decade frequency divider .
So, the LED will be blinking with 40 / 10 = 4 pulses per second ( 4Hz ) ,
when fan rotates at 1200 rpm . ( LED FREQ BLINK = 2* RPM speed / 600 )


And the actual circuit on it's PCB ' module ' - it connects directly to fan - ,
being completed and tested .
Side by side with the SoftOnSet ,delay-relay module ..

 

[robincnn]

Normally Open (NO) is the secret here...
Normally Open Thermostat Switch - 50 Degrees

Thank You

Yes..It happens that I tend to like open-frame drivers...
( * Electrolytic Capacitors : #1 reason of a led driver failing .
Most sensitive parts and prone to
reduction of their service life,as their temperature increases.)
Mean time between failures (MTBF) :
MTBF value prediction is an important element in the development of products.
However, it is incorrect to extrapolate MTBF to give an estimate of the life time of a component,
which will typically be much less than suggested by the original MTBF due to the much higher failure rates in the "end-of-life wearout" part of the "bathtub curve".
http://en.wikipedia.org/wiki/Mean_time_between_failures

Does it mean that the C106 will need to be replaced around every 93036 hours ?

MTBF for HLP is 288k but C106 93,036 hrs , 3 year warranty
MTBF for LPF is 440k but C105 55,161 hrs , 5 year warranty
Demonstrated Mean Time Between Failures is also better for LPF

I wonder what will be more reliable HLP or LPF .



PS: I like the fan blinking thing as long as the blue and red do not flash at the same time.

 

[heckler73]

Don't like electrolytics?
Try polymer
They can take more abuse.

http://www.capacitorguide.com/polymer-capacitor/

That's what Horowitz and Hill use in their LED driver circuit (from AoE 3e).

 

[Doer]

Huh? Learn something from Heck everyday.

 

[Positivity]

lights looking good sds...gets better and better everytime.

 

[alesh]

I believe that all three methods of dimming (MW 3in1) lead to the same result - reduced current.
View attachment 3387197

I emailed them and hopefully they can tell us how it works.

Dear Sir,

Thank you for using Mean Well products.

For almost all of our drivers with 3-1 dimming are reducing o/p current. Only our PWM-90 series ( or other new models in 40/60/120W in PWM-xxx series) is having PWM style output that is suitable for constant voltage LED loads.

Regards
edit: The PWM frequency is fixed @300Hz.

 

[robincnn]

Could not help but notice that it does not have as many air vents as other commercial lights. No side air vents and top vents are small holes. I am curious to know it provide enough cooling for the fixture.

 

[stardustsailor]

Could not help but notice that it does not have as many air vents as other commercial lights. No side air vents and top vents are small holes. I am curious to know it provide enough cooling for the fixture.

Of course it has side-vents to exaust the warm air ..
5,5 mm dia holes ,are not exactly "small holes" ...
>300 of them ,makes a total inlet surface same as the fan's rotor surface area .

Providing enough cooling ?
LOL !



What about this :

4x Vero 29 at 1530 mA have a Tj (junction temperature ) of Ta +25°C !
And that with dissipating about 2 W of fan power ,at <19 dB.
?????

I do not think that are many lights (commercial or DIY ) outta there ,
that can even come close that cooling efficiency ...

Show me ,please ,another example of a LED fixture -with active HSF cooling ,
having a thermal resistance " heat sink to ambient " of 0,14 °C/W !!!

Cheers.

 

[AquariusPanta]

Of course it has side-vents to exaust the warm air ..
5,5 mm dia holes ,are not exactly "small holes" ...
>300 of them ,makes a total inlet surface same as the fan's rotor surface area .

Providing enough cooling ?
LOL !



What about this :

4x Vero 29 at 1530 mA have a Tj (junction temperature ) of Ta +25°C !
And that with dissipating about 2 W of fan power ,at <19 dB.
?????

I do not think that are many lights (commercial or DIY ) outta there ,
that can even come close that cooling efficiency ...

Show me ,please ,another example of a LED fixture -with active HSF cooling ,
having a thermal resistance " heat sink to ambient " of 0,14 °C/W !!!

Cheers.

How does one go about measuring junction temperatures, Sailor? I haven't measured my prototypes for Tj but if it's a feasible procedure, would enjoy jotting down some numbers and the sort.

 

[Positivity]

Never question the cooling masters skills....

Just ask how...lol

 

[stardustsailor]

How does one go about measuring junction temperatures, Sailor? I haven't measured my prototypes for Tj but if it's a feasible procedure, would enjoy jotting down some numbers and the sort.

amongst few ways ,is to use a thermocouple ...
For example ,regarding the Vero 29 ,one can just measure the Tc of the COB with a thermocouple and then
just calculate this Tj = Tc + ( 0.13 * ( 1 - ( If * Vf * eff% ) ))..
That is a rather practical way ...
But it gives also a pretty accurate result.

Cheers.

 

[robincnn]

Never question the cooling masters skills....
Just ask how...lol

I agree. I went through the whole thread and did not see side vents. Will go through entire thread again

 

[robincnn]

Of course it has side-vents to exaust the warm air ..
5,5 mm dia holes ,are not exactly "small holes" ...
>300 of them ,makes a total inlet surface same as the fan's rotor surface area .

Providing enough cooling ?
LOL !



What about this :

4x Vero 29 at 1530 mA have a Tj (junction temperature ) of Ta +25°C !
And that with dissipating about 2 W of fan power ,at <19 dB.
?????

I do not think that are many lights (commercial or DIY ) outta there ,
that can even come close that cooling efficiency ...

Show me ,please ,another example of a LED fixture -with active HSF cooling ,
having a thermal resistance " heat sink to ambient " of 0,14 °C/W !!!

Cheers.

Tc Ta Tj notes to myself
Tc is thermocouple point on cob
Ta ambient temp
Tj junction temp

May I know The Tetras's Tc?

0,14 °C/W sounds really good. Best I heard is 0.46 at optic website. Not sure about A51

 

[stardustsailor]

Tc Ta Tj notes to myself
Tc is thermocouple point on cob
Ta ambient temp
Tj junction temp

May I know The Tetras's Tc?

0,14 °C/W sounds really good. Best I heard is 0.46 at optic website. Not sure about A51

Firstly a small correction :
It's Tj= Tc + ( 0,13* Vf *If * ( 1-eff ) )

Now V series Tetras Tc ,when Ta =25°C :
( Q = total dissipated heat power from 4x Vero 29 )

1) @450mA = 29.75°C ,Rad eff = 47% ,Q= 33.2 W , qua eff= 2.3 umol/J
2) @600mA = 31.55°C ,Rad eff = 46% ,Q= 45.9 W , qua eff= 2.23 umol/J
3) @800mA = 34.03°C ,Rad eff = 45% ,Q= 63.3 W , qua eff= 2.17 umol/J
4) @1000mA = 36.77°C ,Rad eff = 43% ,Q= 82.5 W , qua eff= 2.08 umol/J
5) @1300mA = 40.94°C ,Rad eff = 41% ,Q= 111.6 W , qua eff= 2 umol/J
6) @1500mA = 43.91°C ,Rad eff = 40% ,Q= 132.5 W , qua eff= 1.95 umol/J

As for the side ,exaust vents ...
Voila ...
( Apologies for the crooked pic...)


And the top air inlets ,mostly made out of "small holes" ..
Plenty of them ..Wrist watch is placed as size reference point


Cheers.

 

[robincnn]

Thanks for sharing the numbers and a side pic. Will remember page 14 of this thread for the equation.

This masterpiece is only missing its artists signature logo.