2 transistor asynchronous multivibrator driving logic chip

[Dabbo56]

I have 2 circuits that work no problem:
1. A traditional 2 BJT astable multivibrator generating square wave at 1 Hz (0v to 5v), and
2. A counter 74HC393 chip driven by a function generator putting out square wave at 1 Hz (0v to 5v).
So, I replaced the function generator with the astable multivibrator but getting nonsense output from the chip.
Both square waves look similar on the scope. What am I doing wrong? What do I need to read up on? Thanks!

[Ian.M]

I'd bet the risetime of the function generator is *MUCH* faster than that of your BJT astable, and that the latter is violating the 74HC393's max. input transition rise/fall time given in the datasheet(∆t/∆v, page 3).  The slow risetime keeps the clock input in its transition region where it is sensitive to noise pickup (possibly from its own output transitions) for too long, resulting in extra counts.

Its very hard to get a clean enough squarewave with nice edges out of a BJT astable unless you emitter couple the output.  See attached LTspice sim.

[Zero999]

The transistors in an astable actually turn off very quickly, because the base voltage goes negative. It's the voltage dropped across the collector resistor which makes the rise time sluggish. Add a diode and an extra pull-up resistor to get a faster rise time. Here's an example showing the difference between the extra diode & resistor vs without.

[Dabbo56]

Thanks Zero999 and I’ll definitely implement your modification. One question though, as I understand it the 74HC393 does its thing on the negative going edge which, as you say, happens quickly. Will improving the rise time help?

[Dabbo56]

Thanks Ian.M I'll give it a try

[Zero999]

Thanks Zero999 and I’ll definitely implement your modification. One question though, as I understand it the 74HC393 does its thing on the negative going edge which, as you say, happens quickly. Will improving the rise time help?

It might do. If the dv/dv of the input waveform is too slow, anything could happen. Operating an IC outside of its design specifications is asking for trouble.

[David Hess]

Some companies like Hitachi or NEC put Schmitt trigger inputs on all of their HC logic clock and edge inputs.

[Dabbo56]

The transistors in an astable actually turn off very quickly, because the base voltage goes negative. It's the voltage dropped across the collector resistor which makes the rise time sluggish. Add a diode and an extra pull-up resistor to get a faster rise time. Here's an example showing the difference between the extra diode & resistor vs without.

Yes that worked and the logic chip is now working fine. The rise time which was 68mS is now 34nS.
The fall time is less than 12nS.
Datasheet states max 500nS so well within the limit

[Dabbo56]

As a follow up question, there are 2 ways in which the astable multivibrator circuit can be drawn. In type A the resistor R2 is connected to the base of the first transistor.
In type B the resistor R2 in connected to the capacitor. Both appear in textbooks and references.
Is one better than the other?

[Zero999]

As a follow up question, there are 2 ways in which the astable multivibrator circuit can be drawn. In type A the resistor R2 is connected to the base of the first transistor. (Attachment Link)
In type B the resistor R2 in connected to the capacitor. (Attachment Link) Both appear in textbooks and references.
Is one better than the other?

What's the difference?
Please take another look at the circuits, noting the connections between the components.

[magic]

Type B is better because there is less line crossing and opportunity for confusion

[wasedadoc]

As a follow up question, there are 2 ways in which the astable multivibrator circuit can be drawn. In type A the resistor R2 is connected to the base of the first transistor. (Attachment Link)
In type B the resistor R2 in connected to the capacitor. (Attachment Link) Both appear in textbooks and references.
Is one better than the other?

They are exactly the same circuit.  Just a mater of preference.  Some may prefer the second as it has only one crossover.

BTW.  Running the circuit from more than about 6 volts causes the negative excursion at the bases to exceed the reverse breakdown voltage of many silicon transistors.  That does not necessarily damage them but does cause the circuit to run faster than the usual theoretical calculation. 

[Dabbo56]

Am I crazy? Surely not the same circuit 

[Dabbo56]

Oh I get it. R2 becomes R3

[wasedadoc]

Am I crazy?

Yes.

[Zero999]

Am I crazy?

Yes.

Come on, go easy on the original poster. This is the beginners section.

Oh I get it. R2 becomes R3

There are always different ways of drawing the same circuit. Here's another example. It's exactly the same circuit, drawn a different way.


Wikipedia has an excellent article about this circuit.
https://en.wikipedia.org/wiki/Multivibrator#Astable

[David Hess]

There is also a single capacitor version, shown below.