0
\$\begingroup\$

Up to now, we had 2 COB LEDs ran in series (avoid thermal runaway of UF) driven by a CC power supply (elg-150-c1750b.)

We want to upgrade our setup because we found more efficient solutions in the forms of LED srips (SL-B8T1N60LAWW).

For this exact type I did not find schematics but looking at the parameters and similar strips I suspect the LEDs are used in a mixed arrangement on the strip itself. The strip still does have a sensible "Uf" and typical current defined which I can use for constant current operation.

The min- max operating voltage is 11-13 volts, while other types are 20-25 volts, this kind of suggests these are for parallel operation (12V and 24V CV supplies,) but the datasheet does not clarify the mode of operation.

Is it safe to operate strips with mixed connections using a constant current supply?

We want to use 6 of the mentioned strips in series because normally it is better, but if we have parallel connections involved on the strips, CV drive sounds safer because if 1 LED fails the others parallel with the failed one will not overload.

\$\endgroup\$
5
  • \$\begingroup\$ CV drive sounds safer Sure but these strips appear to be designed for CC (they have no current limiting resistors) so they cannot be used with CV (Constant Voltage)! I would recommend to only use strips in series and take care that the If (current) of the strips is the same or lower than what the driver provides. Also the sum of all the Uf (forward voltage) needs to be lower than what the driver can support. \$\endgroup\$ Commented Mar 3, 2021 at 9:55
  • \$\begingroup\$ If you want to significantly reduce the chance of failure, run the LED strips at a much lower If than they are rated for (for example at half of their rated If). Yes you get less light but you can compensate for that by using more LEDs. \$\endgroup\$ Commented Mar 3, 2021 at 9:59
  • \$\begingroup\$ SL-B8T1N60LAWW looks good. \$\endgroup\$
    – tlfong01
    Commented Mar 3, 2021 at 10:06
  • \$\begingroup\$ SamSung Datasheet: cdn.samsung.com/led/file/resource/2020/09/… specifies CC 1A or 1.5A. So 12VDC with all LED stripes in parallel, within total current limit, should go well. \$\endgroup\$
    – tlfong01
    Commented Mar 3, 2021 at 10:10
  • \$\begingroup\$ The abundance of series resistors sounds logical, thank you for the heads up! we plan to use it around ~1200 mA which is kind of the half of the maximum and and about 200-300 mA less than the typical load. The six strips will also produce a voltage drop adequate for the supply. \$\endgroup\$ Commented Mar 3, 2021 at 10:25

2 Answers 2

0
\$\begingroup\$

enter image description here

These will work well with Meanwell in CC mode with headroom for max current and voltage

There are only thermal concerns that need to be met for a reliable temp rise of 30'C max , so you should not operate more than 1,430 mA @ 55'C unless you manage heat flow.

\$\endgroup\$
5
  • \$\begingroup\$ Thank you for your quick reply Tony! Heat flow is indeed an issue since the LEDs are used in a horticulture project, there is some airflow but the ventilation is pretty weak. The COB LEDs did create some unwanted temp increase, that's why I want to switch to the more efficient Samsung strips. (the COBs had some 130-140 lm/W, not a huge step but still something.) \$\endgroup\$ Commented Mar 3, 2021 at 13:31
  • \$\begingroup\$ Well you have 160W of power consumption per supply. so no getting around that. Unless you use water cooling with your irrigation using a rectangular heat pipe and heatsink compound. (?) Otherwise 20mm wide is barely enough and will burn to touch. Perhaps go to a scrap metal shop and get some 2"x2" wide ALUMINUM channel which can also block glare from the side. like this steel homedepot.com/p/…-PLP_Browse--NA--204225748--N& \$\endgroup\$
    – D.A.S.
    Commented Mar 3, 2021 at 14:15
  • \$\begingroup\$ Another quick, kind of off topic question. I see a luminous flux and a luminous efficacy values. I though if I multiply the luminous efficacy (lm/W) with the power (W) I get the luminous flux, but looking at the datasheet, I do not. The wattage value also does not add up for example for the max values. 13 V x 2,2 A = 28.6 W instead of the 19 W shown in the datasheet. Where does this deviation comes from? \$\endgroup\$ Commented Mar 3, 2021 at 16:47
  • \$\begingroup\$ You cannot apply both limits as a product , either one yes, it not both when the power dissipation exceeds recommended limits based on sense temperature unless you provide sufficient and necessary active coooling. so if you get a bad batch with Vf max you can only use I typ to get Pmax \$\endgroup\$
    – D.A.S.
    Commented Mar 3, 2021 at 20:02
  • \$\begingroup\$ if use I typ, measure Vf or Temp before you consider frying them. A good batch will be more efficient if below Vf typ @ If typ because efficacy always increases with lower current down to 5 or 10% I rated, using less is not suggested by many (metalization) \$\endgroup\$
    – D.A.S.
    Commented Mar 3, 2021 at 20:08
1
\$\begingroup\$

I suspect the LEDs are used in a mixed arrangement on the strip itself.

They are. The load voltage is 11.2VDC (typ), there are 44 LEDs on a strip, so most likely the LEDs are arranged in 4S11P (11 parallel branches of 4 series LEDs per branch) configuration.

Is it safe to operate strips with mixed connections using a constant current supply?

Yes, it is. Looking at the product picture, there are no resistors, so the strip should be driven with CC. The LEDs should always be driven with CC, though.

We want to use 6 of the mentioned strips in series because normally it is better, ...

The load voltage will be ~78VDC max. So you should check the max output voltage of the LED driver that you are planning to use. It should be higher than 80VDC.

...but if we have parallel connections involved on the strips, CV drive sounds safer because if 1 LED fails the others parallel with the failed one will not overload.

Nope. You need a limiter resistor for CV drive. If you drive a single strip with CV, even if only one LED fails (i.e. goes either short or open), the remaining 10 branches are still in parallel, so the drive current (determined by the limiter series resistor) will be divided into the remaining branches which means kinda overload.

So, go for CC anyways.

\$\endgroup\$
4
  • \$\begingroup\$ Surely you have a similar problem when driving a single parallel strip with CC? \$\endgroup\$ Commented Mar 3, 2021 at 11:39
  • \$\begingroup\$ Thank you for your detailed answer, very helpful! How did you get the max 52 V though? We will have 6 strips in series, with a minimum load voltage of 10 V (which is going to be higher I suppose) FYI The constant current region for the supply is 43 ~ 86V so we are good in any ways. \$\endgroup\$ Commented Mar 3, 2021 at 11:51
  • \$\begingroup\$ @TamasVeres sorry. I multiplied it with 4 :) Fixed. \$\endgroup\$ Commented Mar 3, 2021 at 12:02
  • \$\begingroup\$ Thank you for your quick help and clarification! :) \$\endgroup\$ Commented Mar 3, 2021 at 13:24

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.