2

I'm trying to create a suitable response to Pumping a few Amps for a 100usec. The OP wants to create a 5A current pulse for 100us. They want to have a current rise time of 1us at the start of the pulse. This is obviously 5MA/s. Is this likely to cause problems (EMI, ground bounce, etc.) ?

Community
  • 1
Rocketmagnet
  • 26,933
  • 17
  • 92
  • 177
  • Actually, he wants t_rise < 100 ns. – Telaclavo Apr 25 '12 at 11:11
  • @Telaclavo - In his most recent comment, he said <1us would be OK. – Rocketmagnet Apr 25 '12 at 11:14
  • I think it's silly anyway, since taking a picture requires 10 times that time at least, and with flash I don't see any reason to hgo that fast – clabacchio Apr 25 '12 at 11:16
  • Right, I hadn't seen that. He should write that in the header. – Telaclavo Apr 25 '12 at 11:16
  • @clabacchio The way I understand it, the flash will be the infrared light from those LEDs, and the shutter for the CMOS sensor will be open for a longer time. – Telaclavo Apr 25 '12 at 11:20
  • @Telaclavo *shutter :) but still, it's a very quick time, and I don't know what may be so fast to need that speed...And we are talking about rise time, not pulse duration – clabacchio Apr 25 '12 at 11:21
  • @clabacchio 100 us/frame is "just" :) 10 kfps. These guys here take 1 Mfps http://www.youtube.com/watch?v=QfDoQwIAaXg – Telaclavo Apr 25 '12 at 11:39
  • @Telaclavo what I mean is that it's a corner case, and I don't think it's done in infrared too. Second, I don't see the point in having a quick rise time, as opposed to a short duration. The squareness of the pulse shouldn't be that important – clabacchio Apr 25 '12 at 11:55
  • @clabacchio - If he is trying to squeeze as much light into the shortest possible time, with the upper limit set by the LED, then the rise time is important. The total power is proportional to the area under the current graph. – Rocketmagnet Apr 25 '12 at 12:12
  • @Rocketmagnet I don't really know how to interpret this situation, but it looks overkill (unnecessary) to me; it may be lack of understanding of the problem by me or (eventually) by the OP – clabacchio Apr 25 '12 at 12:41
  • @clabacchio - It sounds like overkill to me too. But we don't know enough about his application to know. – Rocketmagnet Apr 25 '12 at 12:48
  • 1
    @clabacchio since you don't understand the reason don't think it is silly. I asked the question, I may not be a great EE designer but I am a pretty smart guy and it is not silly. You just don't know the facts. – Ktc Apr 25 '12 at 15:57
  • @Ktc I realized that it was excessive, but (even if I'm not an expert in photography) I don't understand the real need of an edge so steep for a flash...do you mind about discussing it in a chat? – clabacchio Apr 25 '12 at 16:12

1 Answers1

4

If you assume rise time is near half cycle of your upper bandwidth of radiant EMI on your signal path, tR = 1 uS puts you into the AM band near 500KHz so it may be audible between stations on an AM radio, so FCC CE emanation concerns unless shielded or use twisted pairs and CM choke.

Tony Stewart EE75
  • 1
  • 3
  • 54
  • 182
  • Great comment Tony. In the layout, I will make sure I sandwich this line between ground lines. – Ktc Apr 25 '12 at 15:59