Yes, I've used them. I think they're still available. Some had internal buffering and termination, others were just the delay line. An easy way to tell them apart, given the physcial device, is that the unbuffered ons show a low DC resistance between the output pins (they are taps on a coil inside), the buffered ones do not (they have amplifier circuits between the taps and the pins). Yes. Basically, any transition on the input pin causes a similar transition on the output after the specified delay. So if you feed in a 40ns-wide pulse, say +ve going then and look at, say, the 50ns output : The input is a 0 It goes to 1. 40nx later, it goes back to 0 and stays there after a further 10ns (i.e. 50 ns from the start), the rising edge has 'got to the output'. The output goes high. And after another 40 ns (i.e. 90ns from the start), it goes low again. ---- - ---------- Input ---- ------ ---- Output (50ns delay) Each '-' represents 10ns -tony