Is it possible to phase shift a signal to make use of multiple scope channels?

[lscw]

I'm probably missing out technical terms that better describe what I'm trying to say but here goes:

Is there a reliable way/is it even possible to utilise two or more oscilloscope channels (assuming max sample rate/ch remains the same), each phase shifted appropriately to increase the total sample rate of a signal?

I'm picturing in my head a 2 channel 1Gs/s/channel scope sampling a single signal at 1Gs/s on channels 1 and 2, with a 1ns delay on the signal input to channel 2 and the waveforms overlayed on the display. I guess interpolation might get in the way but is there any other reason it couldn't work?

[elektrolitr]

This is already implemented in many lower-end scopes, when you either have for example 1GSa/s if you use only one channel or 500MSa/s for each if you use both

[filssavi]

What you want is a scope with multiple timebases, while this was fairly common in the old analog dual channel CROs; I don’t know of any modern digital scope (up to midrange such as lecroy hdo8000 or tektronix MSO58) that has it.

Usually with modern scopes you just capture the first waveform once, store it as a reference and then trigger on the second one, that you can delay as much as needed with regular time base control

[Kleinstein]

Combining 2 (or more) ADCs to get a higher sampling rate is a common practice in DSOs, also at the high end. One usually does the delay / phase shift with the ADC clock and not the signal itself. For times in the low ns range,it only takes a piece of PCB trace to get a delay.

Doing this externally is tricky, as high frequency signals need a defined load / line impedance. So already splitting the signal in 2 takes more than a BNC T.

[magic]

Is there a reliable way/is it even possible to utilise two or more oscilloscope channels (assuming max sample rate/ch remains the same), each phase shifted appropriately to increase the total sample rate of a signal?

I'm picturing in my head a 2 channel 1Gs/s/channel scope sampling a single signal at 1Gs/s on channels 1 and 2, with a 1ns delay on the signal input to channel 2 and the waveforms overlayed on the display. I guess interpolation might get in the way but is there any other reason it couldn't work?

I suppose you could, with some preamp/buffer/splitter according to circumstances and a pair of coax cables of different length.

It will make no difference because 1GS/s presumably means "an entry level DSO" and entry level DSOs are limited to a few hundred MHz by their analog preamps, not the ADCs.

And you probably meant 0.5ns delay, not 1ns. The latter is completely pointless.

[bdunham7]

https://www.analog.com/en/analog-dialogue/articles/interleaving-adcs.html

[lscw]

Thanks very much for all the replies. They answered my question directly, and the Analog Devices article make for interesting reading