VC3165 Frequency counters

[rhb]

Does anyone have any comments on these?  They're ~US$80 on eBay with a variety of brand names including blank.

[hexreader]

I own one....

Plus points:   Cheap

Minus points:  Complete garbage

 

EDIT:  I see other forum threads that suggest that VC3165 is OK at higher frequencies (on the B input), which I tend to believe. I mostly use a counter at lower frequencies with the A input, and I find it annoying. The frustration prompted me to spend a lot more money on a BK Precision 1823A, which is not perfect, but a pleasure to use.

[HalFET]

You can get an old HP 5316A for the same price if you're willing to wait around a bit. Or do you really need to go above 1 GHz?

[vk6zgo]

They have an incredibly "clunky" UI, but once you work that out, they are quite useable.
Much better than no frequency counter at all.

[rhb]

I'm in the "no counter at all" state as my old,  low end Optoelectronics  counter is dead from battery leakage in storage.  Most or all of my gear is dead of old age. So I'm looking for cheap new gear that will get me going again in a small space.  No travel anticipated.

The reviews on eHam were pretty good.  Not sure why I didn't think to do a general web search on it earlier.  It's an $80-90 counter in a world where counters can cost as much as a new car.  You get what you pay for, but looks as if it will do for now.

I find it a bit curious that we obsess about ppm accuracy and build things using mostly parts within 5-10% of the required value.

[rdl]

This popped up in my automatic ebay search results last night. It may be more than you want to spend, but used ones show up often for not much higher cost than the Chinese cheapies.

https://www.ebay.com/itm/Tektronix-CMC251-frequency-counter/332455536659

[edavid]

This popped up in my automatic ebay search results last night. It may be more than you want to spend, but used ones show up often for not much higher cost than the Chinese cheapies.

https://www.ebay.com/itm/Tektronix-CMC251-frequency-counter/332455536659

Despite the Tektronix name, that's a very basic counter made by Escort.  It's far inferior to an HP 5315/5316.

OP, if you ever measure audio frequencies, you definitely want to get a reciprocal counter.

[rhb]

I ordered an "Intelligent"  VC3165  on eBay for $81 with free shipping from a US location.  From the YouTube reviews I watched, the interface is horrible, but I think I can live with it.

Not quite sure what a "reciprocal counter' is or why that would matter to audio frequencies.  There is a fundamental theorem which states that frequency resolution is inversely proportional to measurement period.  One can do better than a naive measurement, but you can't escape Shannon.

My Zoom H4N samples at 96KSa/S for hours.  So I can get arbitrarily accurate measurements with that and a bit of programing.  It would be very difficult to build a source that stable. In fact, the clock stability of the H4N would show up pretty quickly.

[edavid]

Not quite sure what a "reciprocal counter' is or why that would matter to audio frequencies.

It measures the period of the input, and computes the reciprocal.  This allows shorter gate times for audio measurements.

There is a fundamental theorem which states that frequency resolution is inversely proportional to measurement period.  One can do better than a naive measurement, but you can't escape Shannon.

I'm sure Shannon knew that you can measure period (and frequency) to arbitrary precision in one cycle of the input.

[rhb]

There is the *minor* issue of having an infinitesimally fine resolution of time.  Measuring one period also assumes that the next is exactly the same length.  Very unlikely to be true in the real world.

You can actually do far better by sampling the waveform and doing an L1 (least summed absolute error)  fit.  That avoids the need for a super fast timer clock.  But it does require a non-trivial amount of computation.  TANSTAFL

Shannon worked out how much data is needed for a given amount of information at a certain level noise (error). While Shannon was focused on the data capacity of a communication channel, his work has broad application in many areas.  For example, mathematically the phase can be derived from the amplitude spectrum alone on physical grounds.  There is, however, a seriously non-trivial issue of how much information is needed.  A VNA can measure phase at a single frequency, but you cannot derive the phase from a single amplitude measurement.  I don't yet know how many amplitude measurements are needed to derive the phase to a given accuracy, but I'm quite sure that Shannon's work applies.

[David Hess]

Reciprocal counters like the HP 5315/5316 do not usually measure the period of just one cycle.  Instead, the user sets the gate time (infinitely variable on an HP 5315/5316) or number of desired digits (Racal Dana 1992) and the period of multiple cycles is measured.

The HP 5315/5316 is a fantastic counter.  The only one I have which is better is a Racal-Dana 1992 but I prefer to use the HP for routine applications.