Ninjabowler
Well-Known Member
Whats that stuff that looks like chewing gum thats sealing around those wires in pic 3?
Analyzing The "SKETCHY" Blackstar 240w - Electrical Engineer Checking In w/ pictures
- Thread starter ragetakesover
- Start date
Ninjabowler
Well-Known Member
Oops 4....pic 4
Not sure, it looks like glue or something else. It seems the workers smear this stuff to fix the wires and switchs. I have been wondering how these products pass the CE and RoHS approval...
guod
Well-Known Member
stardustsailor
Well-Known Member
...and it is used to insulate the switch contacts.Otherwise the appliance cannot get CE and RoHS approval...
But ...
wait a minute ...
All these electronic equip ..
And you didn't knew about the glue ?
Weird ...
But ...
wait a minute ...
All these electronic equip ..
And you didn't knew about the glue ?
Weird ...
Haven't used any such stuffs yet. I believe they have better method to get the wires and switches connected.
![XXO]G81R}U3BE7$[5VQY[~J.jpg](/data/attachments/1734/1734534-b77b64f8153016a1dd903ce9f76a7288.jpg)
stardustsailor
Well-Known Member
Nope....
In Europe insulating Mains Power voltage cables with PET/ PP / Silicone sleeves will not approve to get the CE,at certain cases ...
( open cooling grills ,appliance working in moist environments )
Voltage is 220-240 Volts and they ask for permanent,non-movable, water tight, insulation ...=> thermal glue,silicone, or epoxy covered ..
And more ...
To get the IP xx "splash -proof " or "water -proof " or even "immersion proof " approval,
also sleeves do not pass ...
Neither, for "explosion -proof " approval ....
(direct contact with explosive gas or liquid that might be conductive to electricity .)
So,mostly in/for Europe ,this type of insulation is used in most of cases ...
Along with sleeves also ,in many of them ....
(probably ,they might have sleeves ,those on the panel also ... underneath glue ... )
Neither thermoretractable sleeves pass alone ,in many cases ...(heat makes them soft ..A hot cable can soften them ,or make them retract more ,leaving possibly ,naked cable exposed ...
Anyway ....
Europe has pretty 'strict' and 'absolute' rules ,regarding electrical appliances and approval ...
In Europe insulating Mains Power voltage cables with PET/ PP / Silicone sleeves will not approve to get the CE,at certain cases ...
( open cooling grills ,appliance working in moist environments )
Voltage is 220-240 Volts and they ask for permanent,non-movable, water tight, insulation ...=> thermal glue,silicone, or epoxy covered ..
And more ...
To get the IP xx "splash -proof " or "water -proof " or even "immersion proof " approval,
also sleeves do not pass ...
Neither, for "explosion -proof " approval ....
(direct contact with explosive gas or liquid that might be conductive to electricity .)
So,mostly in/for Europe ,this type of insulation is used in most of cases ...
Along with sleeves also ,in many of them ....
(probably ,they might have sleeves ,those on the panel also ... underneath glue ... )
Neither thermoretractable sleeves pass alone ,in many cases ...(heat makes them soft ..A hot cable can soften them ,or make them retract more ,leaving possibly ,naked cable exposed ...
Anyway ....
Europe has pretty 'strict' and 'absolute' rules ,regarding electrical appliances and approval ...
stardustsailor
Well-Known Member
No...As far as I know this applies for 180-240 AC 50 Hz ..
(for splash -resistance & water-proofness is required at all voltages...I think ...)
(for splash -resistance & water-proofness is required at all voltages...I think ...)
stardustsailor
Well-Known Member
OHHHHH!!!!!
WOW !!!
World's first self-analysing led panel ....
.....
...
..
Way complicated ,uh ?
WOW !!!
World's first self-analysing led panel ....
.....
...
..
Way complicated ,uh ?
Hosebomber
Active Member
That's actually dielectric silicon RTV... SDS is correct. It helps keep the connections together, air/water tight, and reduces risk of corrosion when in humid/wet conditions.
ragetakesover
Member
thanks good reply. please continue to discuss as i may be wrong,
Yes, reverse mounting is functional but efficiency wise it is not acceptable. reverse mounting increases production cost in respect to time (maybe not so much in china). Holes for the LED need to be cut via machine, forward mounting you do not. Machine can position LEDs in both situations. Reverse mounting, i've never seen it done, but seems more troublesome to re-flow the solder (BTW these solder joints look to be done by hand, notice the large solder joints).
Also when revere mounting in this fashion, thermal paste needs to be applied by hand to each of the LEDs. Thermal paste is an inferior heat transfer to a very thin layer of solder. When placed on a metal core PCB, the PCB acts as heat sink and immediately increase the surface are the LEDs heat slug. Although when placing a "metal core pcb" on a heatsink, much more thermal paste needs to be applied in relation to reverse mounting.
Then when placed onto the heatsink using reverse mountig, many more screws are needed to hold the PCB securely to the board so all LEDs make firm contact with the heat sink.
Im sorry I cant find the 1oz copper core. Please explain ...
Your right, the foward voltage is not being ran at max so junction temp should be low. With 3 fans blowing on it, it should remain within safe operating range.
yes there is enough vents for air to escape. Although, the design is still inferior in terms of efficiency to a "pull" configuration. Air being pushed perpendicular to the heatsink creates friction as air needs to be pushed out the vents (ideally speaking the air is being evenly distributed along the area inside the unit, majority of the case is enclosed, not all air is able to escape).
ragetakesover
Member
classic Chinese LED down lights (like your 6" recessed cans in your kitchen), although they stuck 6 of em into an enclosure.
one i have is 11 watts warm white and does not require a fan (gets little hot to the touch).
AC/DC driver - bad - seems to be generic, improper insulation for 120v
Optics - good - most likely 30 degrees on each LED (probably looses 15% efficiency, but intensity on plant is greatly increased)
LEDs - ?? - sorry cant tell which ones, they are not visible
Fan- meh - again a push design, should be a pull design to increase efficiency a little
heatsink - good - finned aluminum heatsink on each unit, with fans cooling it off
overall seems to be a good design for a directional lamp, just crappy components. defiantly would be a good starting ground to upgrade components.
any questions?
ragetakesover
Member
same as above
metal core PCB is forward mount which is the preferred method. (pic 3)
LEDs are generic chinese 1w or 3w. (pic 3)
resistors are used for over current protection. (pic 3)
optics seems to be plano convex, hard to tell by the picture, but they are to focus the light. (pic 5)
ragetakesover
Member
this seems to be a well made light.
lower 4 LEDs are osram golden dragon (pic 3) upper red LEDs, sorry i cannot see clearly enough to tell .. good
forward mount LED on "metal core PCB" .. good
"metal core PCB" mounted onto a large heatsink ... good
only a single 4 watt fan pushing air in (again i prefer pull) ... good
sorry cant see well enough the circuitry but looks to be a pretty cool design, 120v going straight into the PCB (as you can see the arrow with the "high voltage") going into a 4 prong black chip which is a bridge rectifier (lower right corner of pic 1) and into a series of voltage regulators then into the LED array.
cool design ^ ^
Hosebomber
Active Member
Contrary to your thoughts, it is acceptable. Production cost is actually reduced via this method. Generally speaking (excluding prototype boards), nearly all PCBs are machine cut and drilled. Leaving the PCB in the machine for an additional 10-20 seconds to drill the diode mounting holes cost a few cents. Having an additional machine or person to place them that much more precise is expensive. As you noted these are not reflow mounted. They appear to be mounted using the old fashion belt system where the boards ride on a belt above a molten pool of soldier and air blows small amounts of soldier onto the joint areas. This tech has been in use since at least the 80's when I visited the Motorola factory.
Again, the amount of thermal paste used in this system is less than most others. It is applied with a stencil in a very thin (>1 mil) coating in less then a second as the "heat sink" goes down the line. It's precise, thin, clean, no waste, cost and time effective. Most SMD LEDs cannot have the thermal slugs soldiered and most of the cheaper Chinese LEDs do not have an isolated thermal slug at all. Meaning that if they did soldier it, the entire heat sink then becomes a source for a short in the electronics. You stated the problem with MCPCBs and the thermal transference there, so I won't go over that.
As for more screws, that is not the case at all. They are assembled with preset torque drivers. This makes the pressure across the whole board equal and therefore equal pressure on each slug to the heat sink
.
This is pretty basic electrical engineering stuff. 1oz copper core refers to using a 35 micro meter copper trace route for the circuitry on a PCB. The common sizes are 1, 2, 3 and 5 oz. referring to the weight of the material it would take cover 1 square foot. The resulting thickness after covering that 1 square foot is the thickness measure of the traces. Between each dielectric layer (called the core) an additional layer of thin copper is placed for secondary and tertiary routing.
Again, you may want to refresh your basic thermodynamics text. When unducted air is forced over an area of higher temperature, it creates a vortex that rolls the air down over the heated element and back up (to the top of the casing in this instance) then back down again. As the air rolls over the heated surface, transference of heat from the heat sink to the air occurs heating the air and reducing the temperature of the material. This continues to occur until the air finds the path of least resistance to escape, at which time it readily dissipates the heat to the surrounding air. The pull configuration is only marginally better IF the heat sink has fins that the air is forced through. Reading through the wiki on heat sinks will help you immensely.
Eraserhead
Well-Known Member
Fans bowing onto the heat sinks in most cases is better than having the fan "suck" the heat off of the heat sink.
This is why:
If the fan is blowing out, the air first needs to go through the side vents (hopefully the vents are large enough), then over the heat sink, then through the fan.
If the fan is blowing onto the heat sink, it is getting direct air, then the air leaves through the side vents. It's overall less the air has to travel to do its thing.
If you are trying to pull heat off of the heat sink, that means you need to let it get hot first.
By blowing the fan at the heat sink, it prevents it from getting hot in the first place, thus the LEDs will run cooler and are more efficient.
With that said, I have tried reversing the fans on many, many panels, and every time fans blowing in wins.
But also, regardless if the fans are blowing in, out or sideways, if the vent holes are too small or not enough of them, the LEDs are most likely running hotter than they should and will die out right around the time the 1yr warranty runs out (Blackstar for example).
This is why:
If the fan is blowing out, the air first needs to go through the side vents (hopefully the vents are large enough), then over the heat sink, then through the fan.
If the fan is blowing onto the heat sink, it is getting direct air, then the air leaves through the side vents. It's overall less the air has to travel to do its thing.
If you are trying to pull heat off of the heat sink, that means you need to let it get hot first.
By blowing the fan at the heat sink, it prevents it from getting hot in the first place, thus the LEDs will run cooler and are more efficient.
With that said, I have tried reversing the fans on many, many panels, and every time fans blowing in wins.
But also, regardless if the fans are blowing in, out or sideways, if the vent holes are too small or not enough of them, the LEDs are most likely running hotter than they should and will die out right around the time the 1yr warranty runs out (Blackstar for example).
FranJan
Well-Known Member
The thing with panels like Blackstars are those generic LED cases actually add little to the cooling of the diodes. We did some "tests" around here (I'm too lazy to find the link), one weekend where we put some probes on the inside of the cases and on the heatsinks and tested the temps with the fans blowing in and out. In the case of the Blackstar 240w it didn't matter if the the fans were blowing in or out, temps were basically the same. This to me shows the uselessness of the BS's case design to cool the LEDs and the electronics, though the original cases had metal fan guards that when removed really helps cool the inside. To me the fans on a BS are really there to move a little air and cool the drivers which on cheap panels usually put out mucho hot air, especially the guys using those transformer/IC Chip drivers.
This makes me think what Winder says in "Power Supplies For LED Driving"
"When several devices on a circuit board generate the heat, a solution could be to use a
fan to blow air across the circuit board. Cooler air from outside the equipment can be
blown over warm components to reduce their temperature. Airflow will reduce the
effective thermal resistance of the air interface.
Careful placing of cooling fans can make a big difference to the performance. Large
objects like electrolytic capacitors will tend to block the flow and may steer t he cooling air
away from areas o f t he PCB. If the air flows in the direction of heatsink fins, it will be more
effective. Air flowing across the fins will only cool the front and rear fins, see Figure 14.7.
If mounting a fan at the top of equipment make sure that the air flows upwards,
with the fan blowing air outwards, so that the fan aids the natural buoyancy of the
hot air. Fans mounted in the side of equipment are much more effective if two
fans are used, one on either side of the enclosure. In a wide enclosure, both fans
could be mounted on the rear panel; one fan should blow in and the other should
blow out so that air circulates around the components inside.
Fans do have a reliability issue, so consider adding a fail-safe mechanism in case the
fan fails to operate. A fail-safe mechanism should monitor the temperature of
sensitive components on the circuit board. Driving the LEDs at a lower power or
turning them off when the temperature rises too high may be a solution."
Of course he's a bit driver centric but in the case of the BS240w it's the drivers that put out a good chunk of the overall heat. You could probably move the drivers out of the case and cool it with one 80mm fan. Or even remove the driver's plastic case though that could be a little dangerous with those steel cases. ZAP!
This makes me think what Winder says in "Power Supplies For LED Driving"
"When several devices on a circuit board generate the heat, a solution could be to use a
fan to blow air across the circuit board. Cooler air from outside the equipment can be
blown over warm components to reduce their temperature. Airflow will reduce the
effective thermal resistance of the air interface.
Careful placing of cooling fans can make a big difference to the performance. Large
objects like electrolytic capacitors will tend to block the flow and may steer t he cooling air
away from areas o f t he PCB. If the air flows in the direction of heatsink fins, it will be more
effective. Air flowing across the fins will only cool the front and rear fins, see Figure 14.7.
If mounting a fan at the top of equipment make sure that the air flows upwards,
with the fan blowing air outwards, so that the fan aids the natural buoyancy of the
hot air. Fans mounted in the side of equipment are much more effective if two
fans are used, one on either side of the enclosure. In a wide enclosure, both fans
could be mounted on the rear panel; one fan should blow in and the other should
blow out so that air circulates around the components inside.
Fans do have a reliability issue, so consider adding a fail-safe mechanism in case the
fan fails to operate. A fail-safe mechanism should monitor the temperature of
sensitive components on the circuit board. Driving the LEDs at a lower power or
turning them off when the temperature rises too high may be a solution."
Of course he's a bit driver centric but in the case of the BS240w it's the drivers that put out a good chunk of the overall heat. You could probably move the drivers out of the case and cool it with one 80mm fan. Or even remove the driver's plastic case though that could be a little dangerous with those steel cases. ZAP!
Okay, fair enough... So how about you point people in the direction of a good LED manufacturer... I hear 5 watt LEDs will be rolling out soon - any info?
ragetakesover
Member
Thanks, a lot of info you dropped. I did learn some cool things. I have no knowledge of older manufacturing processes. That's why Im trying to open this discussion to gain something. Though the process you stated is now acceptable to me and looks to be a slightly less crappy product.
For our discussion, I still see it as more time consuming and fragile than printing directly onto a metal core.
With metal core, Ive only seen an entire electrically insulated printed sheet on top of the entire metal core. With reverse mount, the slug are in direct direct contact with the heat sink if thermal paste is used (possibility of creating thermal short with non insulated LEDs).
In this case the screws are not evenly distributed between the board, creating increased pressure along certain solder joints. Yes probably the joints wont come undone if left alone inside the unit, although if tampering with the board, the joints will rip the copper paths off the substrate. You cannot deny this is much less stable than direct printing onto metal core pcb.
Ok, cool, I know what your talking about now. Just confused me, I though you were referring each thermal slug containing 1oz of copper weight. Yea the pathways contain no heat because that is were current is flowing through, part of the LED that emits heat is directly under the actual LED chip. No argues on that.
Im sorry lol i still disagree in the thermodynamics portion. Yes i have no argument saying the push effect of a fan will cool the heatsink. Although, what im saying is the pull effect will be more efficient, yes in this case probably be minimal amount of gain as there is enough vents to allow air to flow out of. Even how you explained it says push effect is less efficient in this case. The bouncing up and down of air creates sort of a cancellation effect with new air coming in putting resistance on the fan blades (not saying anything significant here, very minimal), inverting the fan to a pull configuration, i doubt there will be an humanly audible difference as it does not pass the 3db point because the efficiency is not doubled here. Although if you take a db meter you will see a slight decrease in db meaning an increase in airflow.