0
\$\begingroup\$

I couldn't find 3.5796 MHZ Crystal Oscillator for making a DTMF decoder so, I bought a 4.0 MHZ Crystal Oscillator.

Will it have any problem using the 4.0 MHZ instead of the 3.5795 one?

\$\endgroup\$

closed as unclear what you're asking by Leon Heller, Daniel Grillo, pipe, Neil_UK, Dave Tweed Oct 5 '16 at 13:34

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • \$\begingroup\$ Impossible to say! \$\endgroup\$ – Leon Heller Oct 2 '16 at 15:29
  • \$\begingroup\$ anything is possible but a 10% higher base frequency might not be convenient for divide ratios. Both f XTAL's are stock at all major distributors \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Oct 2 '16 at 15:30
  • \$\begingroup\$ Actually possible. Implementation based on MT8870D requires 3.579545 Mhz. \$\endgroup\$ – Anonymous Oct 2 '16 at 15:31
  • \$\begingroup\$ Do you need a crystal or an oscillator? You say oscillator, but a simple crystal is more likely, as a DTMF decoder IC would probably have the oscillator circuit inside it, and need only the crystal and likely some loading caps. \$\endgroup\$ – Chris Stratton Oct 2 '16 at 17:30
  • 1
    \$\begingroup\$ The only way it'll work is if you use the wrong crystal on the matching encoder too. Then they can talk to each other but nobody else can listen in... \$\endgroup\$ – Brian Drummond Oct 2 '16 at 17:39
4
\$\begingroup\$

Assuming that you are using a DTMF chip (there is no information in your question) then the chip will use timing derived from the crystal oscillator to operate the switched capacitor filters in the chip. For example, the MT8870D datasheet says:

Separation of the low-group and high group tones is achieved by applying the DTMF signal to the inputs of two sixth-order switched capacitor bandpass filters, the bandwidths of which correspond to the low and high group frequencies. The filter section also incorporates notches at 350 and 440 Hz for exceptional dial tone rejection (see Figure 3). Each filter output is followed by a single order switched capacitor filter section which smooths the signals prior to limiting. Limiting is performed by high-gain comparators which are provided with hysteresis to prevent detection of unwanted low-level signals. The outputs of the comparators provide full rail logic swings at the frequencies of the incoming DTMF signals. [Emphasis mine.]

In addition, page 10 shows one frequency for the crystal / clock with no deviation!

DTMF crystal

Will it have any problem using the 4.0 MHZ instead of the 3.5795 one?

It certainly will. The passbands of the filters will be out by \$ \frac {4.0}{3.5795} \$. Time to order the right crystal.

\$\endgroup\$
3
\$\begingroup\$

There are plenty of oscillator modules made with that frequency. Digikey lists a dozen different part numbers in stock from four different manufacturers.

If the other end is expecting standard DTMF frequencies, you should just buy one of them (or use an appropriate crystal, but presumably you have some reason for not wanting to do that).

As you can see in the datasheet Transistor posted, the tolerance acceptable is +/-0.1% and your suggestion is +11.7%, or more than 100x worse.

\$\endgroup\$
2
\$\begingroup\$

The DTMF decoder chip uses the crystal frequency to set the detection frequencies. If the crystal frequency is not correct, the decoder will not detect the standard DTMF frequencies.

You need to use the crystal frequency specified by the chip manufacturer.

As a side note - you can use this requirement to your advantage. One project that I did many years ago required that we generate low-frequency signals. It turned out that we could use an off-the-shelf DTMF encoder chip that uses the same crystal as your decoder. This gave much better long-term stability than the original R-C timing network used in the original design.

FWIW - that crystal frequency used to be one of the most-readily available crystals. That's because it was used in every color TV set manufactured to receive NTSC signals.

\$\endgroup\$
1
\$\begingroup\$

It is very unlikely to work. Since you did not reveal what chip you are referring to, we can only offer a generic response. It is rarely beneficial to omit important details in your question.

3.579545 MHz crystals appear to be readily available for a few cents. So cheap that shipping will likely cost several times the cost of the crystal. Oscillator modules for 3.579545 MHz are also widely available, not sure where you are looking?

Note that many/most DTMF generator or decoder chips have built-in oscillator circuits, so a simple crystal is all that is required.

\$\endgroup\$
  • \$\begingroup\$ i'm using MT8870D chip \$\endgroup\$ – Unique Kiddo Oct 3 '16 at 3:09
  • 1
    \$\begingroup\$ It is not clear why you think you need an oscillator module? The MT8870D chip has a built-in oscillator circuit. Simply connect the OSC1 and OSC2 pins to a 3.579545 MHz crystal. \$\endgroup\$ – Richard Crowley Oct 3 '16 at 7:45

Not the answer you're looking for? Browse other questions tagged or ask your own question.