Не в сети
- Сообщения: 10369
- Стаж 9 лет 4 месяца
This device is designed to take the TTL flash signal from a Sea & Sea camera (MM II, MM II-ex, SX1000, etc) and use it to control an Ikelite TTL strobe. The same circuit can also be used to control a Nikon strobe (SB102,3,4,5), by the addition of a resistor of 27 kOhm - 47 kOhm from SP to RDY at the strobe interface.
Circuit description (display the circuit diagram while you read this):
A pull-down on the camera X line turns on Q1 (the bar above the symbol indicates that it is 'active low'). This event ignites the strobe via Q2 and asserts a high on the S (start integration) line, thereby powering the 2902 comparator in the camera. When the integrated light level crosses the TTL set point, the camera asserts a low on its quench line (Q), which is transmitted via Q3 and Q4 to the strobe quench line. The totem pole configuration of Q1 and Q3 permits quenching to occur only while X is low or until C1 discharges. The quench event also triggers the CMOS 555 timer, asserting a high on the camera Q line for a few seconds to light the TTL-OK LED in the viewfinder. The circuit is powered from the ready line of the strobe. Capacitors C3 and C4 hold the circuits up when ready goes low after ignition.
R1 + C1 = debounce network, intended to maintain a clean start signal (S) when the X-contact closes.
R2 - original choice was 10 kOhm, but this gave random triggering with new batteries (6V) in unregulated strobes, due to the current drawn by the ready light circuit in the Motormarine II.
C3 - original choice 100 µF, but this made the viewfinder ready LED stay on far too long after the RDY line from the strobe had gone low.
C4 - increased from 100 µF to 220 µF, to make the viewfinder TTL-OK LED stay on for a reasonable time.
R10 + C5 - timing. OK light stays on for ~ 3s, slightly less if RDY goes low.
R5, C2, C6 - Glitch problem. Original circuit had R5=10K and C2, C6 not fitted. The 2902 comparator in the camera tends to output a short glitch when S is asserted. This caused early false quenching, at random, especially with new batteries in the strobe. Reducing R5 to 1 kOhm, in an attempt to swamp the Q line capacitance, reduced the probability but did not cure the problem. Adding C6 made the mono too slow to trigger on the glitch, so that OK lights in strobe and camera disagreed when premature quenching occurred. Adding C2 cured the problem. C2 delays the Q signal by ~ 1ms, which is not significant. C6 is probably no longer necessary, but is useful diagnostically.
555 - usa a CMOS timer. Prototyped with Texas TLC555P.
When the strobe batteries start to droop, triggering will at some point become unreliable due to the extremely tight voltage headroom of the circuit. Replacing the silicon PN 1N4148 diodes (Vf~0.65V) with Schottky signal diodes such as the IN6263, IN5711 (Vf~0.2V), or the old-fashioned Ge-Au OA47 (Vf~0.15V), will give a small but useful improvement to the voltage headroom.
Artwork and text © D. W. Knight. 1999, 2002, 2004.