Eventually you'd run into the source (wiring) resistance. But that would only be in the extreme.

Another possibility could be if the manufacturer has programmed in a random delay, so that they spread themselves out in time at startup, to lower the cummulative peak.

This is just idle speculation. Standby for others...

Another possibility could be if the manufacturer has programmed in a random delay, so that they spread themselves out in time at startup, to lower the cummulative peak.

That was a technique used during the 7400 series technology heyday when switching times grew faster and faster. Had a product in production using a lot of that technology that originally passed EMC with flying colors to start failing radiated EMC while in production..

Turned out the faster switching time in same part number later vintage 7400 chips was the culprit.. and attempted fix techniques such as larger value chip decoupling caps exacerbated the radiated signal (more energy available at the peak).

After extensive investigation and testing it was discovered some 7400 series chip manufacturers did nothing to compensate for the faster switching times and a few others built in the fix... ring counter like delays within the modules to spread switching times out in time a small amount to lower the cumulative peak switching current draw.

Perhaps that same technique is being used in the subject drivers....

Who said sez it's not accumulative?

Keith Cress
kcress - http://www.flaminsystems.com

One driver manufacturer stated inrush was not cumulative.

After doing further research I looked at NEMA 410 - Performance Testing for Lighting Controls and Switching Devices. They describe the testing procedures for testing inrush. It is cumulative and looks like from Table 2 that the determining factor of how many drivers can connect to a 20 amp circuit is the energy produced within 2 milli-seconds. So I can put together a calculation based on steady state current of my drivers and call the driver manufacturer for the pulse width and get my energy level and compare with Table 2. If I am below the energy values for a certain steady state current, I'm good.

Here is a free download to NEMA 410 document:
http://www.nema.org/Standar:ds/Pages/Performance-T...