74ACT04 VS 74HCT04

Try reading the data sheets. The ACT parts have lower propagation delays and their output stages can sink and source more current. This may or may not make much difference to your tunable oscillator depending on the frequency range over which you wish to tune it.

--
Bill Sloman, Nijmegen.

Try reading the data sheets. The ACT parts have lower propagation delays and their output stages can sink and source more current. This may or may not make much difference to your tunable oscillator depending on the frequency range over which you wish to tune it.

--
Bill Sloman, Nijmegen.

And why is he not reading datasheets?

John

Someone already pointed out, but seems to be consistently ignored... an '04 is TERRIBLE for oscillators since it is three stages of inverter.

'U04 should be used.

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
|  Analog/Mixed-Signal ASIC\'s and Discrete Systems  |    manus    |
|  Phoenix, Arizona            Voice:(480)460-2350  |             |
|  E-mail Address at Website     Fax:(480)460-2142  |  Brass Rat  |
|       http://www.analog-innovations.com           |    1962     |
             
I love to cook with wine.      Sometimes I even put it in the food.

The ACT is a lot faster, and will result in a more stable oscillator because the frequency will be less affected by the active circuit parmeters.

Tam

Try:

The first hit explains the difference:

They differ in speed and power consumption, and yes you can interchange them in many applications, but you might be better off using an unbuffered gate e.g. 74HCU04 for an oscillator.

Why are you trying to make a tunable LC oscillator using logic gates???

Ian

Not sure. May need 74ACU04. Inside the chip 74HC... will do about

400MHz. The big problem is getting signal thru a package pin at that speed.

...Jim Thompson

-- | James E.Thompson, P.E. | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona Voice:(480)460-2350 | | | E-mail Address at Website Fax:(480)460-2142 | Brass Rat | |

| 1962 | I love to cook with wine. Sometimes I even put it in the food.

Thank you all for your comments. I wish to make a square wave oscillator in switching mixer applications so a sine wave oscillator is not even wanted.

Will a 74HCU04 be suitable for a 100MHz oscillator?

JJ

No, The max 1->0->1 delay time is 45 ns. That is 22 MHz. If you are copying somebody else's circuit, stick with what he has ( and the AC part is really marginal), else, use a discrete VHF FET transistor. Lots of circuits in the Radio Amateurs Handbook.

Keep in mind that no free runnong LC oscillator is going to be all that stable at 100 MHz, although it will be good enough for a normal FM receiver.

Tam

In article , Tam/WB2TT wrote: [...]

This will only matter much if the LC circuit's frequency is highish. At low frequencies, it can be the output section's impedance variation with temperature that matters. All such effects are likely to be much less than the tempco. of the inductor involved.

Also: Beware that the output side of either of these chips will look a lot like a squarewave. The peak-to-peak of the fundamental component in this squarewave can be larger than the Vcc the gate runs on. You don't want the input of the gate to be driven to the point where the ESD protection diodes get biased on.

With both of these parts (ACT more so) I would also worry about the stray capacitance in the inductor. If you are trying to make a lowish frequecy oscillator, the circuit could tune up on a combination of the traces and stray capacitances and take off at a high frequency instead of the intended one. This can be made less of a problem by using the "U" version since the "U" version has a decreased gain when you start to get significant phase shift.

--
--
kensmith@rahul.net   forging knowledge

In article , Ian Bell wrote: [...]

Why not? I've made a VCO this way and it worked ok and needed very few parts.

--
--
kensmith@rahul.net   forging knowledge

The HCU part's output won't look very square at 100MHz, but they will oscillate that high. Use the SMT part not the thru-hole one. If you must go thru-hole, don't use a socket.

Don't forget to bypass the supply leg with a capacitor zero inches away. Supply and ground are part of the oscillator too.

--
--
kensmith@rahul.net   forging knowledge

[...] [...]

I've had HCU04s oscillating at 100MHz. The output looks more like a distorted sinewave than a squarewave. I think the makers tend to be conservative about the delay numbers.

BTW: 2N2222 works as an oscillator at 100MHz too.

--
--
kensmith@rahul.net   forging knowledge

Don't mess around trying to screw the last dreg of bandwidth out of TTL

- build your oscillator with ECLinPS parts, hooked up as PECL is you can't afford a negative rail, and use an ECL-to-TTL or PECL-to-TTL converter to get TTL for external use.

--
Bill Sloman, Nijmegen

Great advice, thanks Ken. I'll try the 74AHCU04 if I can find one. Then I'll give the old colpitt transistor osc. a try. What would you recommend to square up the 100MHz waves?

I'm afraid spec sheets aren't so useful when trying to make digital devices behave in oscillator fashion but I have looked at all the relevant ones for all those concerned.

JJ snipped-for-privacy@green.rahul.net (Ken Smith) wrote in news:dprtd0$fa6$ snipped-for-privacy@blue.rahul.net:

I recommend that you don't try to make square waves. Chances are you don't really need them. They may look nicer on a scope but the high frequencies involved are often more trouble than they are worth.

If you want good looking waveforms at these frequencies, you should be thinking of ECL and controlling the impedances of your traces.

--
--
kensmith@rahul.net   forging knowledge

snipped-for-privacy@green.rahul.net (Ken Smith) wrote in news:dpttn2$kj$ snipped-for-privacy@blue.rahul.net:

Ok I won't make them square. What sort of mixer IC's do cell phones use?

JJ