How to generate a 50 Mhz clock for TTL circuit

A listing of Oscillator manufacturers:

Hi,

Thank you very much for your answer. I have the SG-51 and SG-531 @ 50.00 Mhz

Question: If I wire the terminals V_DD to 5 volts, the GND to ground then it will generate a clock signal on the OUT pin of 50 Mhz to make a TTL counter work at that frecuency? (Sorry I haven't wired because I'm afraid to burn it)

Best Regards

What do you want that's different from a modestly priced "canned"

50MHz xtal oscillator?

A resonator, an unbuffered single inverter and a resistor could be cheaper.

Yes, assuming it's designed for 5.0V (not 3.3V or whatever) and assuming the TTL counter can clock that fast.

For example, an SN7490 might only be able to clock at 32MHz, a 74HC74 has a max toggle frequency of only 25MHz @~5V over temperature etc. Fast "TTL" logic families can do 50MHz easily.

Great, thanks for the helpful information

The TTL is a 74HC4040 on page 5 of product specification

shows +25° C @ 82 Mhz @ 4.5 Volts, so I think the chip won't run so hot at 50 Mhz

Best Regards

Wouter van Ooijen

-- ------------------------------------

Webshop for PICs and other electronics Teacher electronics and informatics

Hi again

The crystal oscillator was wired, the signal is sinusoidal. How do I make the signal a square pulse to clock the TTL?

Regards

etc

(Google is your friend, as they say).

Ken

It doesn't matter that the signal looks sinusoidal. I assume the output oscillates between 0V and Vdd (at 50 MHz)? correct?

The output will still cross the TTL transition voltages (or in your case the HCMOS transitions) at 50 MHz +/- the tolerance of the oscillator you purchased, with the jitter spec. of the oscillator you purchased.

Your counter will run at 50 MHz.

--
Yes, it will, _but_ if he's using the output of the oscillator to run
anything else in parallel with the counter, skew might kill him since
Vth will vary from chip to chip.

Check this out:

http://www.semiconductors.philips.com/acrobat_download/various/HCT_USER_GUIDE.pdf

Quite true, John but I didn't see what scope he was using to look at his signal. My Tek 475 at full bandwidth would give a reasonable display but a slower scope would round off his squarewave. I've been using a 32 MHz canned oscillator that makes quite nice squares, considering the 475 can only show the harmonics up to 250 MHz. GG

How do you know? To realy appreciate the squareness of a 50 MHz signal you need at least a 200 MHz scope, more is better.

Wouter van Ooijen

-- ------------------------------------

Webshop for PICs and other electronics Teacher electronics and informatics

Setup: FLUKE 192B Scopemeter 60 Mhz 500 MS/s

The output oscillates as follows:

Scope: Channel A 200 mV - 10 ns Peak Max: 0.63 V Peak Min: -0.24 V Peak-Peak: 0.84 ~ 0.91 V

Multimeter: V_out= 2.476 Volts

Great. I was on the verge of buying an LMV7219

to build the circuit to get the square wave.

Thank you and best regards

Hi,

The clock signal is only for the counter, the counter will feed other TTLs

Regards

Setup: FLUKE 192B Scopemeter 60 Mhz 500 MS/s Channel A = 200 mV - 10ns

That means that the signal is square but as I have only a 60Mhz Scope, it is only possible to see a sinosoidal signal?

Not only possible, but pretty near assured.

page 30 sect 7.4 "it is good practice to terminate all unused LSTTL inputs to VCC via a 1.2 kR resistor. Inputs should not be connected directly to GND or VCC, and they should not be left floating."

I never knew that before.

martin

"An eye for an eye makes the whole world blind" Gandhi

Of course. Basic signal theory.

Wouter van Ooijen

-- ------------------------------------

Webshop for PICs and other electronics Teacher electronics and informatics