3
\$\begingroup\$

I'm a lighting technician that works in the UK film industry. I was really hoping some one could help me with a SSR switching issue.

I have built a led lightning flasher to simulate storm lightning for a film. This fixture will be attached to a drone and will be switched on and off repeatedly for up to 6 minutes in very short bursts due tho the lack of heat sink. I am using 250 W voltage booster and Constant current driver to power two 100 W LEDs these draws 5 A at 30 V. I have 10 drivers and 19 LEDs in my fixture. I will be using two SSRs to switch two separate circuits via a wireless dc controller system. One circuit will switching a load of 40 A and the second will be switching a load of 30 A.

I'm hoping to use a DC-DC SSR like this.https://docs-emea.rs-online.com/webdocs/10c9/0900766b810c9bee.pdf

I have learnt that I need to use a Schottky diode to protect my SSR. I am not an electrical engineer and I really need help in choosing the correct Schottky diodes for this project and where to put them. I have already killed two SSRs after 10 or so flashes and can't afford to kill any more!

I have attached a basic drawing and a photo of my build to help explain what I'm trying to do.

enter image description here LED lightning array

enter image description here

This is the basic set up for driving two leds.

\$\endgroup\$
13
  • 6
    \$\begingroup\$ The best way to do this would be to use LED drivers with enable inputs. They are already switching devices, and they have inductance; the best way of switching them is as a fundamentally integrated part of their design. \$\endgroup\$
    – Chris Stratton
    Commented Jul 7, 2018 at 16:04
  • 4
    \$\begingroup\$ This just doesn't seem the right way to solve the problem. Also you're unnecessarily flying heavy SSR modules. The switching you need is already implicitly present in the LED drivers which are themselves rapidly pulsing switches, you just need to find ones which bring that contol out to an IC pin and then a board terminal. \$\endgroup\$
    – Chris Stratton
    Commented Jul 7, 2018 at 17:44
  • 2
    \$\begingroup\$ The problem with undocumented budget electronics is that unless you find the data sheet for the main IC and reverse engineer the circuit, you don't really know what they do. For this reason questions like this get closed under the "use of consumer electronics" category. \$\endgroup\$
    – Chris Stratton
    Commented Jul 7, 2018 at 19:14
  • 2
    \$\begingroup\$ @JasonWells ah, ok, yeah, I'd recommend not buying electronic components from amazon's marketplace; getting components with specs you can trust and with documentation is usually worth it ;) But here: If you can, try to find the markings on the IC which seems to be half-hidden by the large greenish toroid coil core. That should be your boost converter controller. These things generally tend to have an enable pin, or at least, you can pull up their feedback pin and thus disable them. What type of IC is that? \$\endgroup\$
    – Marcus Müller
    Commented Jul 8, 2018 at 4:44
  • 1
    \$\begingroup\$ @MarcusMüller Apparently it is very common converter U024V aka TE636, ASIN B019W4C5IE. Did not find anything on a chip. Here are some good shots: one, two. 10-pin, fixed 150kHz \$\endgroup\$
    – Maple
    Commented Jul 8, 2018 at 17:56

2 Answers 2

Reset to default
1
\$\begingroup\$

I agree with @tony-ee-rocketscientist and @chris-stratton that your schematics is sub-optimal for the job, due to current surge on input capacitors of DC-DC voltage boosters.

You can look for voltage boosters with "enable" input and use that to trigger LEDs, as @chris-stratton suggested. This would be simple off-shelf solution but it does require some shopping.

Or you can move SSRs from before boosters to between boosters and LEDs. This way the SSRs won't be affected by input capacitance. I am not sure I share @tony-ee-rocketscientist worry about output capacitors in this configuration. You can always add bigger caps on the outputs.

What I worry about is how the CC drivers will behave when load is disconnected. Will they jack up the output voltage too much? Only experiment can tell.

\$\endgroup\$
1
  • \$\begingroup\$ The drivers have an adjustable current and voltage limit, just like a bench power supply. So disconnecting the load shouldn't be an issue (if he has the voltage limit set correctly). I've used these drivers several times myself. \$\endgroup\$
    – Drew
    Commented Feb 10, 2021 at 20:51
0
\$\begingroup\$

Your problem is not fly back protection which is already built-in to the SSR but rather peak current driving the 100uF input cap which has an ESR <<0.1 Ohm which creates a I pulse = short circuit current of LiPo’s. >200A ?

FWIW, The pulse width must be equal to 1 picture frame or so.

This surge current might be prevented using the SSR on the output of the booster but then the output caps may fail from RMS ripple current over time. Adding bigger caps will make it more robust but then Voc must be tuned correctly.

So it is a sub-optimal solution due to the storage cap pulse currents in this CC,CV design in pulse mode for that DC-DC converter SSR configuration.
A better cheaper design is possible but without, use the SSR on the LEDs directly with addition cap on output such that Vout is optimized and 1 ohm series R’s are added to balance each shared LED Vf to share current equally as I have outlined in previous posts here (search my answers )

\$\endgroup\$
10
  • \$\begingroup\$ Hello Tony, thanks for taking the time to reply to my question. I don't think the SSR does have fly back built in. heres is a drawing for the SSR that i want to use./Users/JasonWells/Desktop/Screen Shot 2018-07-07 at 17.43.58.png . So if I connect the ssrs at the out-put end of the drivers they may survive? This lamp only needs to survive for two nights of use. \$\endgroup\$
    – Jason Wells
    Commented Jul 7, 2018 at 16:46
  • \$\begingroup\$ From the link to SSR they have zener on MOSFET. The documentation clearly requires external diode for inductive loads \$\endgroup\$
    – Maple
    Commented Jul 7, 2018 at 16:50
  • \$\begingroup\$ Hello Maple, please could you suggest a external diode that would do the job please. As you've probable guessed I'm not a electrical engineer. Any help would be very gratefully received. Kind regards Jason \$\endgroup\$
    – Jason Wells
    Commented Jul 7, 2018 at 16:54
  • \$\begingroup\$ @JasonWells As Tony says, what's most likely kills your SSR is not a flyback voltage (DC-DC boosters are NOT inductive loads, they are capacitive) but a surge of current when you switch LEDs on. I doubt diode will help you much \$\endgroup\$
    – Maple
    Commented Jul 7, 2018 at 17:00
  • \$\begingroup\$ Hi Maple, I understand now thanks. The SSR that died was rated at 60amps, on a 30 amp load, I stupidly thought that would be safe, obviously not. So if I install a 100amp SSR would that survive?? As Tony suggested maybe placing the SSR on the LED side of the booster may work better for longer. As I said it only needs to work for two evenings, so a long life is not required. The LI-Po has a discharge rate of 150amps with a max burst rate of 300amps. \$\endgroup\$
    – Jason Wells
    Commented Jul 7, 2018 at 17:41

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.