CASIO fx-3600P scientific calculator from 1985

[jebem]

(05-31-2018 05:47 PM)Csaba Tizedes Wrote:  OK, as I can understand, the frequency is depends on the reciprocal of the resistor value: f~1/R.

That is my understanding too, but the actual formula will depend on the R*C time constant plus additional constants and operators, depending on the oscillator circuit in use.

Because the oscillator is internal to the chip, there is on way to know in advance what exact formula to use.
That is why people playing with this subject of overclocking use the experimental method and sometimes one can guess the actual formula from the test results and even guess the internal oscillator circuit in use.

To summarize, the lower the R*C time constant, the lower the oscillator period and higher the frequency.

(05-31-2018 05:47 PM)Csaba Tizedes Wrote:  Meantime I tried to find a calculator which I can dismantle and I think I will try it on a 3400P if it has same resistor inside or I have one workable but not so good shaped CASIO fx-6300G. If inside I found a crystal oscillator, can I try to change it?

To be honest, I am not a adept of overclocking anything anymore (those days are gone for me), although I enjoy to see other people fiddling with it.


Concerning the 6300G:
Yes, I vaguely remember to see a resonator/crystal inside one of my 6300G machines and this would give a more precise oscillator frequency useful to operate the graphical LCD properly.

Another resonator could be used, but this may cause issues on the LCD operation. As I said before, one would need to experiment. It is a try and error job.


As for the 3400P/3600P:
I do not have the 3400P in my collection, but being a low end series like the 3600P, my guess is that you may find a couple of resistors and a capacitor inside as well.

The low value resistor is usually used to limit the current consumption of the calculator and should not change the oscillator frequency.

The high value resistor (above 100Kohm typically) is the one to check.
There is no need to remove the resistor. Just solder another resistor in parallel.

Concerning the additional resistor value to use: It is a test and try job.
- The starting point would be around the same value as the one installed from factory, giving a final resistance value of half of the original value.
- Depending on the test results, other resistor values could be tested. If the machine hangs, use a resistor with at least two to four times the original value and try again. If it works, use lower values until it stops working.

The capacitor, if it is a low value (in the range of 20 to 200pF) could be a candidate to check too, but usually this single capacitor is there to bypass/filter the power supply, so it is not a first option to try. That said, a visual inspection and a multi-meter will help here: if it is connected to the power supply lines, it is not part of the R*C constant.
And if it is really part of the R*C constant, here you would need to remove the capacitor and install a new one with a lower capacitance value (unlike resistors, putting two capacitors in parallel will add to the capacitance value).