3
\$\begingroup\$

I have a device (a strobe light) which draws up to 15 A for up to 50 ms at a time, but doesn't allow duty cycles over 0.1. This means that the average current is at most 1.5 A.

For instance, if you drive it for 10 ms, it will ignore all inputs for the next 100 ms.

The question now is: how do I choose a power supply for this light? I assume allocating 15 A for this device alone is a major overkill, but how do I know if weaker power supplies will be able to keep the voltage at 24 V during the 50 ms strobe time? Other devices depend on this voltage as well.

The second question is which circuit breaker do I choose for this device? C2 breakers should be able to handle 2 to 20 A for short amounts of time, but how do they handle repeated bursts?

The strobe light is the Smart Vision Lights' ODSXP30: http://smartvisionlights.com/downloads/datasheets/ODSXP30_Datasheet.pdf

The power supply I intended to use is the Weidmuller's CP SNT 250W 24V 10A: http://catalog.weidmueller.com/catalog/Start.do?localeId=en&ObjectID=8708680000

Assume other devices besides the strobe light can use up to 8 A.

Edit:

To clear up some points from the comments:

The 50 ms pulse duration limit is enforced by the strobe controller integrated into the light. It also enforces the duty cycle limit and therefore the average current limit. That would mean that the max RMS current is 4.75 A.

The 15 A pulse is limited to 50 ms by the strobe controller integrated into the light. The controller also enforces the 0.1 duty cycle / 1.5 A average current limit.

The power supply is a pretty much standard simple DIN rail AC/DC converter for 220/24 V.

Here is a quick simplified schematic:

schematic

simulate this circuit – Schematic created using CircuitLab

\$\endgroup\$
8
  • 2
    \$\begingroup\$ If you have constant current limiting on the power supply output, you can easily just add a suitable capacitor. If not, you probably need a resistor in series to not cause OCP, but this varies from manufacturer to another. Calculate the RMS current and check the time constant for the fuse if way below it, you will be fine. If on the edge, you need to check a A^2t graph for it. \$\endgroup\$ – winny Jan 17 '17 at 13:58
  • \$\begingroup\$ @winny Thanks for the tips, I just have a couple of follow-up questions. How do I know if the power supply has constant current limiting? Do I use RMS, average or max current when calculating the required power supply current? I will be using circuit breakers, not fuses, so is it enough that it can handle a single pulse to guarantee it can also handle them repeatedly or do I still need to calculate RMS like for fuses? \$\endgroup\$ – relatively_random Jan 17 '17 at 14:49
  • \$\begingroup\$ If it does not say so explicitly, it does not have it. Most LED supplies have it. Depends on the time constant. If your peak is second long, you need to calculate for peak current. If your capacitor-resistor or capacitor-constant-current arragement have longer time constant than your load, you can pretty much calculate for average. This goes for both dimensioning the power supply and fuse unless I missintrepreted where your fuse goes. Can you draw a schematic? \$\endgroup\$ – winny Jan 17 '17 at 15:08
  • 1
    \$\begingroup\$ @winny The strobe controller is not sensitive, but other devices probably won't like a drop of more than a volt or so. An RC circuit crossed my mind as well, but it would probably not be very small considering it needs to be able to take almost 5 A RMS. Also, I'm wondering if it would affect the stability of the power supply in any way. \$\endgroup\$ – relatively_random Jan 17 '17 at 19:42
  • 1
    \$\begingroup\$ I understand that. If you can tolerate a bit less than 24 V to your strobe controller, you can calculate an RC solution. If your other loads can accept a bit lower voltage, you can calculate a constant current-capacitor solution. EDIT: Then you are in a bit of a tight spot. Output capacitance isn't usually rated for commercial power supplies. Can you accept a diode for each load, capacitor for each and output 24.7 V? \$\endgroup\$ – winny Jan 17 '17 at 19:43
2
\$\begingroup\$

You are in a tight spot here with one supply and two differnt load requirements. Your obvious brute force solution would be a 23 A supply. If that's not a solution for you, I can see three ways forward, each with their own pros and cons.

Constant current limited power supply:

schematic

simulate this circuit – Schematic created using CircuitLab With 200 mF and 5 A available for the strobe, you will limit the drop to 20 V during a 50 ms 15 A burst. If this is ok with your other loads, you are good to go. I'm suspecting this will be an expensive and bulky solution.

Normal 24 V power supply plus RC filter:

schematic

simulate this circuit If you can spare 4 A peak and live with ~20 V average on your strobe, this will work with a normal constant voltage power supply. A bit wasteful though.

CVCC supply with individual diodes:

schematic

simulate this circuit

Both other loads and stobe will see slightly lower voltage than 24 V during charging but if you can tweak your supply to slightly above 24 V to give you some headroom plus compensate for the diode drops, this might be ok.

\$\endgroup\$
0
\$\begingroup\$

How about a boost/flyback converter? When the strobelight is off, the Converter goes into very low duty cycle operation, as not to rise above the 24 volts.

When the strobelight is ON, whatever your input voltage is (12 volts available?) gets stepped up to approx. 24 volts, with a bit of ripple.

I've used something like this, decades ago, to step up 1.5v battery to power a 5volt MCU.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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