Clock Distribution - Flip Flop Question

[bratwurst]

Hi,

First post here.   I'm a software guy but have been hacking away on Arduino's for the last couple of years and have been leaning more and more towards hardware.  I'm also a Ham, so RF projects are what interest me most.   In summary, I know enough to be a menace with a soldering iron but that's about it.

I have a collection of HP test equipment (5245L Counter, 3325A Function Generator, 8640 Signal Gen) and for fun I want to use a rubidium standard (at 10MHz) as my lab standard.    Problem is the 4245L requires 1MHz, the 3325A can use 1 or 10MHz, and the 8640 needs 5Mhz for the external timebase.

I've built a splitter circuit using a 4013 flip-flop (div / 2) and a 4017 counter (div / 10).  They are running at 9v to support the relatively high working frequency (10MHz).  Works great, but I need 3V p-p to trigger the logic gates.   The rubidium puts out 1v p-p.   Reading online, I can open up the rubidium box, and get a 5v p-p square wave, but I'd like to also be able to use the 5245L as an alternate standard, as the crystal oven seems to be fully working and is probably more accurate than I will ever need.   The 5245L also puts out 1v p-p on the 10MHz reference output.

My question is, what approaches can I take to make a 1v signal trigger the logic chips?   I've tried to build a single transistor pre-amp, but my parts box doesn't currently have anything that wants to run at 10MHz.   Before I run off to the parts store, is a pre-amp the only approach? Or is there an alternative I should be looking at?   

My preference is for a solution that involves off-the-shelf discrete components rather than something I'll need to mail-order.

Cheers,

Michael

[ddavidebor]

A schmitt trigger

[ddavidebor]

Also you need to filter out tha square wave to reduce the noise in your lab

[SeanB]

Look how to use CMOS as an amplifier, and use a 4069 as the amplifier, with the other gates being used to clean up the signal to the dividers. As well use a 4049 or 4050 as a buffer to drive the coax after your dividers. You will need to use a resistor of 51R to match the output to the coax, and bypass the IC's with a 100n capacitor across the supply pins, and a 100uF capacitor by the input and output chips. Better for powering is to use a 7815 and run all of them off 15V, best slew rate but highest power use.

[edavid]

Your counter is a 5245L, not a 4245L.

Any ordinary transistor like a 2N2222 or 2N3904 will work OK at 10MHz.  The problem is probably your circuit, so post it.

[bratwurst]

Cool thanks.   A few leads for me to follow up.    Going to play with a transistor amp first, not sure why but my previous attempts gave 0 voltage gain at 10Mhz...

And yes its a 5245L

[bratwurst]

Thanks for your help and hints guys.  I managed to complete my frequency splitter.   The post on another forum said "all you need is a flip-flop and a low-pass filter, that should be easy for an advanced class Ham" (not addressed to me).   As with a lot of posts online, its often a lot easier to type about than to actually do.

I don't have a schematic, but a quick inventory of the modules I needed to build for it to be successful.

- 50 ohm signal splitter for the 10 MHz input (toroid transformer)
- Input pre-amp based on 2N3904
- 1.2 Mhz low pass filter, and 5.5MHz low pass filter for 1 and 5 MHz outs
- Attenuators to bring output on 1 and 5 MHz outputs to 13dbm

Result is clean sine wave on all outs, so something went right.