If a circuit design uses molded ceramic decoupling caps for all the TTL and CMOS logic circuits can you use monolithic ceramic cap or tantalum caps instead of the molded ceramic decoupling caps?

Replacing all the molded ceramic doubling caps and using either monolithic ceramic caps, silver micas, tantalum for decoupling caps would change or cause problems to circuit board?

 

The ceramic capacitors are non-polarized. Tantalum caps are polarized. So yes these may cause problems.

 

The ceramic capacitors are non-polarized. Tantalum caps are polarized. So yes these may cause problems.

What type of problems can it cause and why?
when using non polarized decoupling caps compared to using polarized decoupling caps the circuit would react how so?

 

If a polarized cap is inserted into a circuit with the wrong polarity then the results will be very dramatic and very bad. Tantalum caps explode and give off a very displeasing odor.

The purpose of the ceramic cap is to supply current to the TTL or CMOS logic circuit during high current current pulses. A ceramic capacitor will, as a general rule, have lower impedance at high frequencies than a tantalum cap. Although a silver mica cap might work in place of a ceramic cap, its size will be quite large for the same capacitance value.

 

Often capacitors are used to bolster the voltage to a circuit. It could be that you have electrolytic and ceramic in parallel already! Do you have a picture or schematic?

 

Decoupling capacitors usually are very low cost and not-all-that-great performance. There are several different types of monolithic ceramic dielectric materials, with code names such as X7R and NPO. Tantalum electrolytics are expensive, and there is the possibility of installing them backwards. Silver mica caps are not polarized, but have much better temperature stability and dialectric absorption than needed in a decoupling application.

ak

 

modern subminiature electrolytics do the job for low mhz

 

This PDF has some useful info on caps.

 

Most TTL and CMOS logic circuit boards have 50 decoupling caps. So It's ok to change all the decoupling caps on a circuit board that are ceramic caps to electrolytic or Tantalum? This won't cause any problems or noises, coupling issues?

 

Most TTL and CMOS logic circuit boards have 50 decoupling caps. So It's ok to change all the decoupling caps on a circuit board that are ceramic caps to electrolytic or Tantalum? This won't cause any problems or noises, coupling issues?

it costs more. ttl is energy eater who is really still using them for large circuits?

when you multiplex leds yo also have spikes so need electrolytic anyway and for low frequency no additional ceramic is needed.

 

The ceramic capacitors are non-polarized. Tantalum caps are polarized. So yes these may cause problems.

I think it was IBM that used 3-pin tantalum beads - the 2 outside leads do the same thing, so its impossible to get them the wrong way round.

Tantalum capacitors are very vulnerable to any reverse current - they have been known to go off like match heads!

 

Most TTL and CMOS logic circuit boards have 50 decoupling caps. So It's ok to change all the decoupling caps on a circuit board that are ceramic caps to electrolytic or Tantalum? This won't cause any problems or noises, coupling issues?

No, tantalum or other electrolytic caps are _not_ a good substitute for a ceramic caps in a TTL circuit.

A smaller value ceramic cap actually has _lower_ impedance than an electrolytic cap at the high frequencies of TTL edge speeds. You want low impedance in the cap so it can supply the current for the high current spikes when a TTL output switches state.

 

You want low impedance in the cap so it can supply the current for the high current spikes when a TTL output switches state.

Why would you want the high current spikes when a TTL output switches states?

 

Why would you want the high current spikes when a TTL output switches states?

The totem-pole output stage of TTL *CAUSE* those spikes - you need the decoupling caps to *try* to supply that transient current draw.

 

The totem-pole output stage of TTL *CAUSE* those spikes - you need the decoupling caps to *try* to supply that transient current draw.

Do you know why the totem pole output states cause a current spike? its because the transistors are switching and the decoupling caps smooth out the current spike?

 

the modern kinds are better. controllers can run with 30% voltage variation in a wonky mode with a clocking coil.

so, by controlling the external supply voltage you can frequency of the mcu.

but red leds can dip to less than 2v

 

Do you know why the totem pole output states cause a current spike? its because the transistors are switching and the decoupling caps smooth out the current spike?

TTL and other logic devices cause power supply current spikes because the totem (push-pull) output transistors are on at the same time when the output changes state form high to low or low to high. This is a very shot time but is long enough to create a voltage drop on the power supply and and ground wiring. The bypass capacitor is placed physically close to the TTL part so the length of wiring that can drop voltage is small.

 

TTL and other logic devices cause power supply current spikes because the totem (push-pull) output transistors are on at the same time when the output changes state form high to low or low to high. This is a very shot time but is long enough to create a voltage drop on the power supply and and ground wiring. The bypass capacitor is placed physically close to the TTL part so the length of wiring that can drop voltage is small.

There are whole series of appnotes devoted to "ground-bounce" with recent high speed and low voltage logic families.