Lets talk about Frequency counters

RichardO

Joined May 4, 2013
2,271
This was a qty2 full cesniun atomic clocks similar to those used by WWV (two large consoles). orNIST https://www.time.gov/. They received th 10Mhz signal and did a phase comparison with the 10Mhz signal out of Colorado. you could the ionosphere move causing a phase shift to be recorded on the chart recorder. When we adjusted the crystal oven in the higher dollar equipment we would display the10Mhz standard and measure drift in nanosecond s over a fixed time. this tested both check accuracy and stability.I is odd I never mentioned WWV (the call sign for the
NIST but it is assumed I did, I wouldn't be surprised if the atomic clock output is not used for the10Mhz base freq. It would make sense.AWe used the 10mhz standard factory wide Most godd equipment has an external time base input.
A few years ago I stood close enough to brush my shoulder against NIST-F1 cesium fountain atomic "clock", the primary frequency standard at the time. (No pun intended).

Note that the cesium clock is really not a clock; it is a very precise frequency reference. The actually clocks at NIST are hydrogen MASER's that are periodically calibrated to the the cesium standard.

There are surplus rubidium time standards available:
http://www.diyphysics.com/2012/02/14/d-i-y-10-mhz-atomic-clock-frequency-standard-using-surplus-rubidium-oscillator/

Finally, you can get access to a very accurate time reference using a GPS receiver. Some modules have a 1Hz output. maybe some of these do but I am not sure:
https://www.sparkfun.com/pages/GPS_Guide
 

Thread Starter

Wendy

Joined Mar 24, 2008
21,938
Counters

Counters are the heart of a freq counter. They determine the upper freq response. When I first started w/ electoronic you were doing well to get a CMOS gate to respond to1Mhz, which is laughable by modern standads. A 1Hz time base means a 1Hz resolution. This is the low end. When I started this I was thinking 4 digits.We have one PIC decade couter in Completed projects forum Which I find a shame. Given a 6 digit counter can be uses in so many instruments. I still can't do PICs or arduionos, so I will have to do this old school. During my research for the digital clockI came across the 4029 an4033 two CMOS mod 10 counters w/ built in 7seg display decoders.To be continued...
 
Last edited:

nsaspook

Joined Aug 27, 2009
6,900
After decades in a junk box it's time for the rebirth. Cut the end off an old USB A to B cable and solder the 5vdc wires to the board. Check the power input for shorts and plug it in to a USB port for power.
It seems to know that time as passed (2001, I'm sure I built it in the early 70's) but the time-base is not working correctly (no output).


Found and fixed a few loose solder connection and adjusted the crystal trimmer for the time base.
Crystal-controlled multivibrator with series-resonant crystal.
https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19720000155.pdf

Bingo, we have the correct frequency from the quartz crystal.

That signal goes to the first divider stage of 3 7490 counters (2 in divide by 5 mode, 1 in divide by 10 mode )
for a 1khz output that feeds into another counter stage.


That looks good. The rest of the circuit has a 7400 for the gate functions with a 7474, 7490 and 7492.
Can't remember exactly how that worked so I'll have to trace that circuit later.
 
Last edited:

nsaspook

Joined Aug 27, 2009
6,900
Reversed engineered one of my own devices.:oops:

Basic clock and gate circuit traced.

Adjusted the xtal trimmer for a 1.0ms gate using the TEK period count function.

IC7A3 display counter pulses #1 and IC3 pin 11 1ms gate pulses #2

IC7A3 250khz input #1, gated output signal #2
Input signal (tapped the 250khz clock for testing) and gate.

IC8B8 display counter reset gate #1

IC7B6 Another gate signal #1 that I not sure about it's original use.:confused:

The display reset gate signal wire (yellow/green stripe) is not connected to the display module, it was broken off at a solder joint somewhere on display board in the distant past so tracing that circuit will be the next job.
 

Attachments

Last edited:

nsaspook

Joined Aug 27, 2009
6,900
Finally working again after repairing two broken lands and wires.

counter reset gate circuit. Falling edge input, positive pulse output.

#1 1ms gated signal IC7A3 from frequency counter input IC7A2, IC8A3 #2 pulse output to reset display counter chain.

Detail: IC7A3 output to 7490 counter input #1, IC8A3 counter reset pulse #2

250,000hz input (from its xtal timebase) to output display in khz.

The rest of the circuit is just 4 9490-7447 seven segment counter and display stages with the reset signal IC8A3 connected to pins2&3 of the 4 digit
7490 cascade.
OK, back to the circuit mausoleum where it belongs.
 

nsaspook

Joined Aug 27, 2009
6,900
Hey I just cheated and got a Calibrated Fluke Freg counter lol..

Good luck with the build
That's not cheating but it doesn't quite have the personal coolness factor of building your own primitive version, finally getting it working again many years later and then remembering that the IC6 7474 circuit for the gate signal was a gate synchronizer to eliminate 1 digit bobble from the classic TTL Cookbook .:D
https://web.stanford.edu/class/ee183/handouts_spr2003/synchronization_pres.pdf
 

gdm_klog

Joined Feb 24, 2013
10
I was hoping that this thread would be a little more retro in nature, but it's all modern stuff with ICs, 7-segment output.
Now, in my garage, I have a one decade counter/ display "module" with just 4 tubes for the decade counter and ten neon bulbs for the display.

Now, that's retro... :)
 

RichardO

Joined May 4, 2013
2,271
Now, in my garage, I have a one decade counter/ display "module" with just 4 tubes for the decade counter and ten neon bulbs for the display.

Now, that's retro... :)
In tech school there was an HP counter with the 10 neon bulbs per digit. I think that was the same unit that used phantastron circuits for the divide by 10's in the timebase.

Phantastron:
http://www.radartutorial.eu/17.bauteile/bt52.en.html
 

Thread Starter

Wendy

Joined Mar 24, 2008
21,938
I apologize, I dropped the ball on this thread.Life dragged me down.

The counter

Below is a preliminary schematic. It has a lot of work left to go.It is not the best , I really don't like its inability to count while the screen is blanked AKA latch an hold, but it will work.When I complete the schematic I will post it on this thread given the basic simplicity Dead bug style prototyping might work well.

99999 Counter.png
 
Last edited:

nsaspook

Joined Aug 27, 2009
6,900
I was hoping that this thread would be a little more retro in nature, but it's all modern stuff with ICs, 7-segment output.
Now, in my garage, I have a one decade counter/ display "module" with just 4 tubes for the decade counter and ten neon bulbs for the display.

Now, that's retro... :)
That's not retro, that's prehistoric.:) We had one of the 0-9 HP tube frequency counter units at Tech school.
 

RichardO

Joined May 4, 2013
2,271
I apologize, I dropped the ball on this thread.Life dragged me down.

The counter

Below is a preliminary schematic. It has a lot of work left to go.It is not the best , I really don't like its inability to count while the screen is blanked AKA latch an hold, but it will work.When I complete the schematic I will post it on this thread given the basic simplicity Dead bug style prototyping might work well.

View attachment 129926
I think you will like this circuit. Not very practical but it uses 555 timers. :D
The circuit uses the 60 Hz power line frequency to get a 1 Hz timebase. The 555 timers are used in a manner similar to the phantastron circuits I referred to in post #33.

upload_2017-6-28_15-46-14.png
 

Attachments

Top