What is the least count of the timer clocks used in RADAR? I was checking out some videos in YouTube regarding the working principle of RADAR. To quote some HOW IT WORKS: World War II Radar (720p), part 1, 
How does RADAR work? | James May Q&A | Head Squeeze. 
They mention that the time taken by a radio wave pulse to reflect off from a target is used for ranging or estimating its distance from the RADAR. 
Then the least count of the clocks used in these RADARs has to be very precise. But possibly not existed in World War II. 
This rises the question - How exactly the RADAR works? Is it possible for RADAR to work with 1940s clocks?
 A: "How exactly the RADAR works? Is it possible for RADAR to work with 1940s clocks?" is, of course, two questions, and the second is easier to answer: in the sense that we talk about it today, where a signal is analyzed and a digital readout provides timing information, 1940's radar did not do that at all, and therefore did not "work with clocks" at all.
The first question is very broad, but I'll give it a try. 
Early radars did not use digital displays - they used analog displays based on oscilloscopes. A good introduction is http://en.wikipedia.org/wiki/Radar_display. Basically, a timing circuit produces a voltage ramp which is applied to one axis of a cathode ray tube (CRT), which is what scopes were made with before LCD screens and cheap processors and digitizers came along. This produced a bright dot on the screen which moved so fast that it looked like a line. The ramp was synchronized to the transmit pulse of the radar. The amplitude of the return signal was coupled to the other axis of the scope. The result was a line whose shape showed bumps (target returns) at a position which was proportional to the distance from the radar. From the Wiki article,
 
shows a radar display with the range displayed vertically, and the amplitude of two different beams displayed horizontally (with the amplitudes shown in the opposite directions - zero amplitude runs up the middle, and the left beam moves left for more return, while the right beam moves right). So this display shows a pair of identical returns- one close (at the bottom of the trace) and another farther away (2/3 of the way up), with a third return about halfway up which has a much bigger return on the right than on the left. The range to each return can be read off from the divisions marked on the sides of the display, and each division corresponds to some calibrated distance.
A more classical display, and one which has shown up in movies since after WWII is the Plan Position Indicator (PPI) http://en.wikipedia.org/wiki/Plan_position_indicator

In this case, the antenna rotates, and the ramp voltage is rotated in synchrony. Getting a strong return causes a bright dot on the screen. The screen has a series of concentric rings overlaid on it, and these rings can be used to measure range to the target. In this example, a target is seen at about 1 3/4 rings. If the rings are located at 10 mile intervals, the target is about 17 miles away.
In both of these cases, there is no automatic range measurement, and no clocks are used. The radar operator figures the range by eye, and reports this to somebody else.
By current standards the range accuracy was actually pretty good. WWII naval radars were typically accurate to about 0.1%. So a radar which detected a plane at 20 miles (35,000 yards) might be accurate to 35 yards, or a bit over 100 feet. Reading the trace or plot to this accuracy wasn't very common, and besides, there really was no way to use this accuracy.
A: One does not need a precise clock for radar because the time that a radar system is  measuring is very short, on the order of a few microseconds to maybe a millisecond (that's already 300km of distance!). The long term stability of the timing system is therefor irrelevant, but it's long term stability that makes precise clocks hard to build. 
In terms of distance measurement the WW II radar systems were probably not much better than a few parts per thousand, i.e. they could get the distance right within a few hundred meters. That was good enough because detecting enemy planes, ships and U-boats was far more important than determining their position down to a meter or less (which modern systems can do). Back then they didn't have weapons systems that could target that well, so they didn't need targeting radar that accurate. 
As for the clock and timing sources they used to create precise signal timing, I would take a guess that most of the systems were based on precise temperature controlled or compensated oscillators and delay lines. A delay line is just a piece of cable that's coiled up for a few to a few hundred nanoseconds of signal delay. Longer electrical delays can be made from discrete LC circuits for the  microsecond to millisecond range. You can still buy those from specialty manufacturers who make mostly military grade equipment and parts, although sales should be next to non-existent these days, check out http://micro.apitech.com. I have no idea who still uses these. In physical labs the use of delay lines is still common, but we usually make them ourselves, why pay someone to cut a piece of cable for you, right?  
For the display of the radar signal the most common technique was to use a swept electron beam in an oscilloscope tube, which requires a precise sawtooth oscillator and a gain calibration circuit. Both were most likely calibrated by measuring the distance to geographic objects that were known to the operator. 
That is at least what I would have done with the means of the time. One could certainly have built digital counters with electron tubes that may have worked at a few MHz, but I am not aware that such devices were actually used, even though the quartz oscillator was invented before WW II. As far as I am aware digital technology was not common in these kinds of systems until well into the 1950s, but then I am not a tech historian, all my data points come from schematics and practical work with some mid century military systems (believe it, or not, when I was in the military, we had a radio that was developed during WW II but wasn't put into production until the late 1950s and that was in use well into the 1980s at a time when fully integrated digitally tuned radios were the standard in civilian life, already!). Today all of this is done with phase switching and Fourier transformations, of course... so it's basically a lot of math running on digital signal processing systems. 
