Frequency counter. Do you really need one?

[Citizen]

I am saking myself, why should you buy FC   if every dig. scope nowdays have one build in.
 Yes. Thats a question.

[nctnico]

Frequency counters are often more accurate. I have an old HP5334A and I have used it about 10 times over the past 10 years but when I use it, it serves its purpose well.

[Lightages]

The need for a frequency counter depends on the individual's interest or need. If you need to measure frequencies less than 20-40MHz, there are many multimeters that have this capacity to measure within their display limits and accuracy.

If you need to measure higher frequencies or have higher accuracy or more resolution, then yes there is a need for a dedicated meter.

In other words, if you don't know if you need a dedicated frequency counter, then you do not need one.

[ve7xen]

A 'universal' counter can also measure many other timing parameters of a signal that may be awkward or inaccurate to measure on a scope. For example A->B interval, A/B frequency ratio. It will also have adjustable trigger thresholds and usually the option of using an external gate and time reference. And of course much higher resolution.

And it doesn't require any 'scope channel(s).

[hgg]

Hi, by the way, do you know if you can use a frequency counter like the one below (RF)
http://www.409shop.com/409shop_product.php?id=104535  and convert it somehow
so that you can measure the frequency of signals with higher voltages peak to peak?
e.g. 10V peak to peak.

[dfmischler]

Frequency counters are often more accurate. I have an old HP5334A and I have used it about 10 times over the past 10 years but when I use it, it serves its purpose well.

I have an HP 5315B and it's just about the same deal for me.  At one time I worked for a company that did a bunch of microwave stuff and they had guys with some serious counters and a cesium frequency standard to calibrate them to.  They needed their counters every day.

[jpb]

I've been wondering about this myself. Frequency counters, compared to other equipment, are quite cheap and the ones provided with scopes are limited.

For example, the one on my scope is 6 digits (many are just 5) and is limited by the bandwidth of the scope (350MHz). It is also confined to the trigger channel, though I know other scopes allow you to choose. It is also tied to the accuracy of the scope, there is no external 10MHz input. For a typical scope this may be only a few 10s of ppm. I tested mine with a GPS 1pps signal and it measured it as 999.997 mHz so was within 3 ppm which I was pleased about (my scope timebase should be within 10ppm).

I wouldn't buy an 7 or 8 digit counter as that wouldn't be sufficient a step up, but a 9 or 10 digit counter allows you to check frequencies very accurately if you can make use of an external GPS or Rubidium source. (A counter that doesn't allow the use of an external frequency standard might only be accurate to 1ppm so is the same order as the scope's counter.)

I don't really need such accuracy as I'm not a radio enthusiast, but I'd like to have it. And there are lots of experiments I might do which would be helped by a counter. It also provides a means of checking the timing aspects of other equipment (such as a scope).

[G0HZU]

For servicing of ham/CB radio gear a decent counter is an essential piece of gear to have on the bench because this type of equipment tends to use numerous crystal oscillators that each need to be adjusted for accuracy using a counter that should typically offer 0.5ppm accuracy.

When I was a student I used to use an old Marconi TF2430 counter for stuff like this. Its very basic internal reference is typically within 2Hz of 10MHz after a 30 minute warmup which isn't bad.

However, if you don't need to service radio gear then I think a decent counter probably isn't needed by many people.

I also have a couple of reciprocal counter/timers from Philips and Anritsu and these are very useful for measuring lower frequencies with very high accuracy and very fast update rate and also they can time events very accurately. The old TF2430 can't really do this.

Note that you can buy a decent counter on ebay for very little money if you are patient. I would avoid the cheapo chinese counters on ebay and instead look for the usual vintage makes such as Racal or Philips and always choose one with an external 10MHz option.

[vk6zgo]

I have  a cheapo Chinese counter---I couldn't pass it up for $A99 at the local Hamfest!

Apart from being a bit flimsy,it is good enough for most work I will do,& probably better than some of the very early RACALs & such!
It certainly counts a lot higher in frequency.

My Tek 7613 has a vertical amp output on the back,so if I stick the FC on that output,I can accurately determine the frequency of whatever signal I am probing.

This is useful when checking through Ham Transcievers of which many have multiple oscillator circuits.
Obviously,I could do the same with a DSO with a built in frequency display,but from what I have seen,they have quite limited resolution.

[EEVblog]

In other words, if you don't know if you need a dedicated frequency counter, then you do not need one.

I concur. Although when you do need a good one, there is no substitute.
Some counters have meany features, like actual counting (input pulses) and other stuff which can be handy.

[G0HZU]

I still think there are better bargains for the patient purchaser if you buy a used counter.

eg for £65 I bought a 3GHz Anritsu MF1603A counter on ebay. It came with all the top options including the option 3 oven that has 0.005ppm stability over temperature and 0.0005ppm per day. Also GPIB to allow logging of measurements remotely.

It's probably 17 years old but it was one of the finest (and fastest) counters on the planet in its day. eg it can display 10MHz to 0.1Hz with several updates a second or it can display 11 digits on the slowest gate time. No $99 chinese counter on ebay can compete with that

It's outclassed by the best of today's counters from Agilent or Tek etc but you can pick them up very cheaply if you are patient because Anritsu gear tends to get overlooked in favour of HP/Agilent yet the build quality of the japanese counter is arguably better.

[vk6zgo]

I still think there are better bargains for the patient purchaser if you buy a used counter.

eg for £65 I bought a 3GHz Anritsu MF1603A counter on ebay. It came with all the top options including the option 3 oven that has 0.005ppm stability over temperature and 0.0005ppm per day. Also GPIB to allow logging of measurements remotely.

It's probably 17 years old but it was one of the finest (and fastest) counters on the planet in its day. eg it can display 10MHz to 0.1Hz with several updates a second or it can display 11 digits on the slowest gate time. No $99 chinese counter on ebay can compete with that

It's outclassed by the best of today's counters from Agilent or Tek etc but you can pick them up very cheaply if you are patient because Anritsu gear tends to get overlooked in favour of HP/Agilent yet the build quality of the japanese counter is arguably better.

Agreed,but my $99 counter was handed to me at a Hamfest brand new----no ebay,or postage cost involved.
As it is,I now have a counter,whereas I would still be looking for one,if I went the secondhand route.
I am a great booster of Anritsu,though,having used their Spectrum Analysers,which are extremely well made.

[manticore00]

One exception to cheap(both in price and quality) Chinese imports is probably the VC3165. I wish it had an external timebase input but otherwise, based off of my own experience, and the reviews I read before I bought one, I'm quite impressed given its price point.

[alex77]

One exception to cheap(both in price and quality) Chinese imports is probably the VC3165. I wish it had an external timebase input but otherwise, based off of my own experience, and the reviews I read before I bought one, I'm quite impressed given its price point.

I have bought one a few months ago mainly for measuring high frequencies that my 100mhz siglent can´t reach. I agree with you, I think that its a great tool for the price.

[bingo600]

I got a cheap China one with a lousy Ref-Xtal , but i was lucky as it used a 13Mhz Reference.
And there is a cheap 26Mhz docxo on the *Bay
Get the docxo , and an AC74. And you have a superstable 13Mhz ref.

https://www.eevblog.com/forum/testgear/zhaoxin-hc-f2700l-%28atten-f2700c%29-frequency-counter-un-stable-count/msg205383/?topicseen#msg205383

/Bingo

[Tepe]

One exception to cheap(both in price and quality) Chinese imports is probably the VC3165. I wish it had an external timebase input but otherwise, based off of my own experience, and the reviews I read before I bought one, I'm quite impressed given its price point.

Is it possible to add such an input so you could use an external 10 MHz signal?

[Telequipment]

 I'm fortunate enough to have Two one is  frequency counter, from 5 Hz to 600 Mhz, the other is a universal Counter timer that goes from 0.1 hz to 10 Mhz, both used for my Amateur Radio the !0 Mhz one, is mainly used  with my HF transceiver which doesn't have digital readout .

[bingo600]

I have (by accident) actually used one of my counters to debug some microcontroller programs.
I had a 10ms timer irq that i was watching (toggling a pin) , but noticed that once in a while the frequency went off ....

I later found out i (once in a while) had a stack/heap collision , and the "counter told me about it" :-)

I have lots of them (but i'm addicted ...  time-nut)

The China Thingy with the 26Mhz OCXO

These are driven by 10Mhz from my Tbolt GPSDO.

PM6671
PM6674 - TCXO
HP5384a
RacalDana 1991

And the Ref' one
PM6680B

I actually use the HP384a a lot , for plain measurements.
It's so easy to set up , and is quite accurate. (with Ext 10Mhz)


/Bingo

[PA4TIM]

I have several including a few frequency  references. My first one was a kit with a pic that went up to 55 MHz. the rest varies from two 60 MHz nixie counters upto a 18 GHz counter. I builded a 1GHz counter last year for use in a signal generator but that became to big and 1 GHz was not enogh so now I made a very simple but very acurate one. Nothing more as a MB506 prescaler, a fet, and a 4013 to it get divided by 512. Then a 4060 with watch Xtal and a 4013 to get 1 Hz as gate. An Arduino programmed ATmega328 does the rest. No calibration needed. It was only 3 Hz  high at 512MHz (so in real 3 Hz devided by 512) Did not bother to get is better. But this only reads frequency. Commercial versions can do a lot more. Attenuation, trigger, gatingtime, periode, frequency, delta two channels and some a lot more.

But the answer to the question:
No, you do not need a counter because if you did, you would have know. 

[bingo600]

Then a 4060 with watch Xtal and a 4013 to get 1 Hz as gate. An Arduino programmed ATmega328 does the rest. No calibration needed. It was only 3 Hz  high at 512MHz (so in real 3 Hz devided by 512)
But the answer to the question:
No, you do not need a counter because if you did, you would have know. 

That's quite impressive as i don't expect you to have the watch Xtal@Bodytemperature 

Ohh and i do agree , if you really need a counter ... You know ...

/Bingo

[PA4TIM]

The 32kHz is divided back to 1 Hz so the Xtal ppm/C does not matter much  I played around with N0P caps and a very small ceramic trimmer to get it adjusted at 1000.0000 ms after the devider (a whole turn is about 500ns adjustment) but later I tested the effect and it does not matter. The fault is below 1 Hz and my resolution stops there. The final version reads only with 1 kHz resolution what more then enough is for my purpose (sweep/signal generator upto 2 Ghz) http://www.pa4tim.nl/?p=2662
The schamatics are not yet upto date. I will do that when he is ready. The arduino software is there too, but there is a small bug in it. Not a big thing, i forgot to uncomment a line that closes the string with a zero. I will change that toO.

The meter also measures power

[alank2]

This really is a great ocxo:

http://www.ebay.com/itm/Pletronics-26MHz-OCXO-Miniature-Oscillator-NEW-/130718871982?pt=LH_DefaultDomain_0&hash=item1e6f73b1ae

I use it in my AVR based reciprocal frequency counter.  It is still on a breadboard missing a decent input stage, but works great...

[Marc M.]

Is it possible to add such an input so you could use an external 10 MHz signal?

If the counter uses a 10MHz internal clock then yes you can.  You'll need to add some buffering/impedance matching and there may be some phase issues but for most users they won't matter.  I run my 30+ year old B&K 1801 counter off a rubidium standard piped into its xtal oscillator.  For most stuff I don't need more than 1Hz resolution and if I slip and contact high voltage, it's less than $1.00 US to repair the front end.  Prior to modification it was all over the place, varying in the 10's of Hz and would never stabilize.  I have a HP5334a (also running off the rubidium standard) I use for critical or higher frequency measurements (1801 is about 60MHz vs 5334a is 1.3GHz).

Marc -

[alank2]

I use mine all the time for calibrating things.  It uses one of the Pletronics 26MHz OCXO's from eBay (seen in the lower right).  It is based on an ATMEGA328.  It uses a MC74AC4040N to handle frequencies from 0-100MHz and a MAX7219 for  the LED display.  The picture shows an 80 MHz oscillator.  It has min/max/avg and is a reciprocal style counter.

[jpb]

I've just bought an ex-demo TTi TF930. It is a reciprocal counter with 10 digits and, importantly for me, goes down to 0.01Hz - most counters I've seen on e-bay start at around 10Hz. It also has no dead zone in counting (i.e. the counters aren't stopped when measurements are taken). It also has remote control via USB.

The reason I wanted the lower frequency was to allow me to connect a GPS 1pps whilst at the same time connecting an OCXO to the external reference. I can then adjust the external reference to get exactly 1Hz to 10 figures. A lazy approach to a GPS disciplined oscillator.

So far I'm happy with it. With the internal oscillator it reads 1.000 000 020Hz and it is very steady except when I regularly lose the satellites (I've not yet set up my outdoor antenna).

12 digits would be nice but the TF930 seems a good compromise between cheap 8 digit counters (which I felt wouldn't add much to the 6 digit one on my scope) and nice Agilent ones which even on their e-bay store are over £1000 (the TF930 as ex-demo cost me just under £200 ex VAT). I'd been looking on e-bay for a while and most 9 or more digit counters were quite old, didn't go down below about 10Hz or above a couple of hundred MHz and often were in the region of £140 to £200 anyway. And buying directly from TTi I get a years warranty and I like to support local (to me) companies though my spending budget will not make much difference to their turn-over!

The only negative (so far) is that the display is not backlit so needs good light, but this is partly to keep the power requirement down allowing the counter to be run from its internal rechargeable batteries or from the USB connector.