These LED lights include a buck converter. www.aliexpress... www.aliexpress... www.aliexpress... www.aliexpress... www.aliexpress... tpower-ic.com/p...
It's a shame you didn't measure the current at the same time to get the actual power used at various voltages, that would give a better indication of comparative dissipation. Excellent video though!
One big advantage of the simple resistor+LED strips is PWM dimming, which is a good way to reduce heat and increase lifespan since you can easily under-drive them with a fixed voltage source.
Suggestion: run each kind of unit at the same perceived brightness (use them to light a dark room, adjusting the voltage on the resistor one to match the buck regulated one) and see what current/voltage each drew to see what's more.efficient.
A buck is going to be on/conducting for a lower amount of time when fed with a higher input voltage, which is where the efficiency over a linear dropping style regulator comes from
I would've expected it to run hotter at a higher voltage too with my limited knowledge and I assumed mppt controllers operate in a similar manner however my controller (renogy rover li 60A) has efficiency graphs and for a 12v battery system 40vmp coming in at 100% load is claiming around 96.5% efficient and with 110vmp around 93.5% efficient. I have a 40vmp string and a 120vmp string on identical controllers. The one under load operating around 37v I think was chilling at like 36°c and the controller with 120vmp was at like 61°c both on identical pv strings just series vs parallel
@@faded.0913 This can be approximated like this. Let's say there is 10 watts out and 10 watts in. If vout is 10 volts then current out is 1 amp. If vin is 20 volts then current in is 500 mA. If vin is 30 volts then current in is 333 mA. If vin is 40 volts then current in is 250 mA.
Less current at higher voltages, and of course less current at voltages below 12V. So the current is probably at a maximum at around 12 volts. Interesting :)
Definitely run at higher voltage then you will save a lot of wire guage. Most solar systems are 48 volts as this is pretty safe. If you are not too worried about humidity then 60 volts is also a useful standard. Ideally you need current limiting dc circuit protection and then running at 80 volts because more reasonable. The future of solar home is low amperage protected DC circuits as this is really much more efficient. Also we need to use more aluminium wiring to reduce copper usage globally.
14:43 I think it's good if they are rated to operate up to 85V because automotive electrical system can get very nasty spike coming from electric starter, relay, etc.
the panel with the buck converter - what was it bucking the 12v down to? I'm looking to replace the LED tape in my shed ( not as fancy or complex as yours - just a few 12v LA with Solar for outside lights, bluetooth amp and an invertor to charge power tools )
You have to compare the output in lumens between the old and the new before comparing temperature. If the new are outputting 2X the lumens for 5 degrees C higher temperature, for example, then they are still much more efficient. And yes, the buck converter will run more efficiently at a higher input voltage because the switching frequency will be lower and the total power dissipated through that inductor will be lower, ergo lower temperature.
Why not both? Longer original strips for general light, smaller buck bariant for specific lighting? I couldn't tell how long the original strips are, but if they're significantly longer, they'd get light spread out more per strip.
COT chip on toast? I like them though, LED units with plain exposed aluminum backs are so great because there's so many ways to heatsink them if the huge exposed plate isn't enough for some reason
the huge alu plate is just a heatspreader and NEVER enough to run at full rated current. might be sufficient @ like 25%. i always put bigass coolingfins on every power led project and they just last. heat kills silicon. in CPU/GPU, inverter, amplifier environments everybody knows but leds are being run without heatsink (and fail) all the time. 😂
Not sure which LEDs I'd use, I suppose the ones with the lowest watt per 'lumen'. For the shed standard supply voltage though I'd definitely go for some fixed value, probably 24 (or 48 if I had enough batteries).
🔦 Those look like they could be blinding if you look diectly at them. Recently I bought a brand new LED flashlight and said to myself "I wonder what the led looks like when it is lit." After going blind in one eye for a couple of minutes, that was the last time I made that mistake. Cheers from Canada :-)
I wonder what the EMC performance is like? Some of these cheap LED fittings spew out so much interferencebthat my DAB radio stops working when I switch the lights on. I would only use warm white versions. Those blue white types are harsh and don't do the circadian cycle any favours.....
I'm not sure comparing temperature of one converter vs one resistor gives you the conclusion you've made. Would a better indication would be current? Just from a simple point of view, there are many resistors vs one converter, which your conclusion didnt account for, no?
Now there's the point J - SiBs drop voltage dramtically from ca4V to 2V so If you set up a 24V grid you end up with 12V at low SOC. Its a linear decline. a PITA. So you need a boost converter to compensate say 1000W ca £25 EB. Do these work ok? with a battery bank, are they reliable/overheat on full power? Over stressed components fail in 3 months or less you know the drill? Lets all find out about these as no one talks about this major issue with SiBs
If you're using a chemistry like Sodium Ion with a wide-ranging voltage, it seems like the LED panels with the buck converters are a better idea, so long as you can keep the minimum voltage high enough. IIRC, that's an effective swing from 2.0-4.1 volts, right? (I remember there wasn't much capacity in the 1.5-2v range.) That being the case, any battery bank that's at least 6 cells in series should be fine, and you'll have even lighting all the way down to a "dead" battery. My guess is you'll have other electronics that have other issues with a swing of 12-24v.
I like the new ones. Also, on a totaly different subject, Dan from SimpleElectronics, wants to ask you onto his podcast, but he can't find a way to contact you though email.
@@lorddonpromillo4836 Currently (8 days later, so possibly updated) they work fine from Sweden. I'm a frequent AliExpress shopper though, so might have a bunch of cookies set so might be recognized and handled a bit differently...
It's a shame you didn't measure the current at the same time to get the actual power used at various voltages, that would give a better indication of comparative dissipation. Excellent video though!
One big advantage of the simple resistor+LED strips is PWM dimming, which is a good way to reduce heat and increase lifespan since you can easily under-drive them with a fixed voltage source.
For reducing heat and increasing lifespan it's better to use an external resistor instead of PWMing it. It's not much a difference but it's something.
I've dealt with some of the bluer COB lamps by covering a portionof the surface with Kapton tape to get a more pleasant color temperature.
Suggestion: run each kind of unit at the same perceived brightness (use them to light a dark room, adjusting the voltage on the resistor one to match the buck regulated one) and see what current/voltage each drew to see what's more.efficient.
A buck is going to be on/conducting for a lower amount of time when fed with a higher input voltage, which is where the efficiency over a linear dropping style regulator comes from
I would've expected it to run hotter at a higher voltage too with my limited knowledge and I assumed mppt controllers operate in a similar manner however my controller (renogy rover li 60A) has efficiency graphs and for a 12v battery system 40vmp coming in at 100% load is claiming around 96.5% efficient and with 110vmp around 93.5% efficient. I have a 40vmp string and a 120vmp string on identical controllers. The one under load operating around 37v I think was chilling at like 36°c and the controller with 120vmp was at like 61°c both on identical pv strings just series vs parallel
@@faded.0913 This can be approximated like this. Let's say there is 10 watts out and 10 watts in.
If vout is 10 volts then current out is 1 amp.
If vin is 20 volts then current in is 500 mA.
If vin is 30 volts then current in is 333 mA.
If vin is 40 volts then current in is 250 mA.
Maybe look at your stock of buck converters and see if they all run cooler when reducing more than near the input V. That discovery was interesting.
What's the current draw on the COB lights? I would count that as a significant factor in any battery powered consideration.
Is the hot inductor still a lot less wasted power than the entire board being the same temperature with distributed resistors?
Constant power means less input current when the voltage is higher.
Got the middle one, about 7-8 pieces, they are bloody brilliant, really work on any voltage up to 90v, I think. And they are cheap.
What were the typical current draws at various input voltages? The prices are really low for the strips. Which is a big plus.
Less current at higher voltages, and of course less current at voltages below 12V. So the current is probably at a maximum at around 12 volts. Interesting :)
Definitely run at higher voltage then you will save a lot of wire guage.
Most solar systems are 48 volts as this is pretty safe.
If you are not too worried about humidity then 60 volts is also a useful standard.
Ideally you need current limiting dc circuit protection and then running at 80 volts because more reasonable.
The future of solar home is low amperage protected DC circuits as this is really much more efficient.
Also we need to use more aluminium wiring to reduce copper usage globally.
14:43 I think it's good if they are rated to operate up to 85V because automotive electrical system can get very nasty spike coming from electric starter, relay, etc.
Higher voltage means less current through the inductor = less hot
Double the voltage and it halves the current. It's inverse proportional.
the panel with the buck converter - what was it bucking the 12v down to?
I'm looking to replace the LED tape in my shed ( not as fancy or complex as yours - just a few 12v LA with Solar for outside lights, bluetooth amp and an invertor to charge power tools )
Everybody shouting why didn't you measure the current at 12v and 24v to see if the current drops at higher voltage.
You have to compare the output in lumens between the old and the new before comparing temperature. If the new are outputting 2X the lumens for 5 degrees C higher temperature, for example, then they are still much more efficient. And yes, the buck converter will run more efficiently at a higher input voltage because the switching frequency will be lower and the total power dissipated through that inductor will be lower, ergo lower temperature.
Why not both? Longer original strips for general light, smaller buck bariant for specific lighting? I couldn't tell how long the original strips are, but if they're significantly longer, they'd get light spread out more per strip.
COT chip on toast? I like them though, LED units with plain exposed aluminum backs are so great because there's so many ways to heatsink them if the huge exposed plate isn't enough for some reason
the huge alu plate is just a heatspreader and NEVER enough to run at full rated current. might be sufficient @ like 25%. i always put bigass coolingfins on every power led project and they just last.
heat kills silicon. in CPU/GPU, inverter, amplifier environments everybody knows but leds are being run without heatsink (and fail) all the time. 😂
Looks like also if you increse the voltage the diode gets hotter.
Yes, that became apparent when I was editing the video :)
Not sure which LEDs I'd use, I suppose the ones with the lowest watt per 'lumen'. For the shed standard supply voltage though I'd definitely go for some fixed value, probably 24 (or 48 if I had enough batteries).
Handy for using on 48v solar systems 👍
🔦 Those look like they could be blinding if you look diectly at them. Recently I bought a brand new LED flashlight and said to myself "I wonder what the led looks like when it is lit." After going blind in one eye for a couple of minutes, that was the last time I made that mistake. Cheers from Canada :-)
They look nice I have to say👍🇮🇪
I wonder what the EMC performance is like? Some of these cheap LED fittings spew out so much interferencebthat my DAB radio stops working when I switch the lights on. I would only use warm white versions. Those blue white types are harsh and don't do the circadian cycle any favours.....
Good point. The LEDs with resistors don't oscillate :)
Apply tape to metal surfaces for thermography. Otherwise only reflection is visible.
Yes, paper tape like for wall painting
I'm not sure comparing temperature of one converter vs one resistor gives you the conclusion you've made. Would a better indication would be current? Just from a simple point of view, there are many resistors vs one converter, which your conclusion didnt account for, no?
Now there's the point J - SiBs drop voltage dramtically from ca4V to 2V so If you set up a 24V grid you end up with 12V at low SOC. Its a linear decline. a PITA. So you need a boost converter to compensate say 1000W ca £25 EB. Do these work ok? with a battery bank, are they reliable/overheat on full power? Over stressed components fail in 3 months or less you know the drill? Lets all find out about these as no one talks about this major issue with SiBs
I'm thinking of making an 8-cell sodium battery specifically for these lights. 32V at top and 16V at bottom would suit these lights well.
If you're using a chemistry like Sodium Ion with a wide-ranging voltage, it seems like the LED panels with the buck converters are a better idea, so long as you can keep the minimum voltage high enough. IIRC, that's an effective swing from 2.0-4.1 volts, right? (I remember there wasn't much capacity in the 1.5-2v range.) That being the case, any battery bank that's at least 6 cells in series should be fine, and you'll have even lighting all the way down to a "dead" battery. My guess is you'll have other electronics that have other issues with a swing of 12-24v.
Would you entrust it in your expensive electric car
No. But fine for the shed.
I like the new ones.
Also, on a totaly different subject, Dan from SimpleElectronics, wants to ask you onto his podcast, but he can't find a way to contact you though email.
When any of these are set in shed?
As soon as it stops raining :)
@@JulianIlett Roger, Roger...
I reverted to resistor strip in workshop. Reason being if a resistor fails lose three leds. If the buck converter fails then lose the entire led pad
True. Although the LEDs on the resistor strips seem to fail intermittent and flicker like crazy.
I know that all too well. Pity no mix of the tech
Should be perfectly usable with my EUC (it has 20S li-ion battery - 84V at 100% charge)
Literally none of your product links work.
Worked fine for me a day after posting the video…..
@@Subgunman I think his links are uk specific
@@lorddonpromillo4836 Currently (8 days later, so possibly updated) they work fine from Sweden.
I'm a frequent AliExpress shopper though, so might have a bunch of cookies set so might be recognized and handled a bit differently...