I received my contest prize from Kerm a bit over a week ago. Since then, I've been extremely busy, but I have had the chance to try to overclock it, as described in this topic.

I successfully overclocked it, with potentiometers for maximum adjustability, but there was one problem. Originally, the calc ran at 6Mhz in speed mode 0, and 16 in mode 1. When I added the potentiometers, those speeds fell to around 4 Mhz and 11Mhz, which I feel is unacceptable.

The decrease in speed is due to an increase in capacitance. This increase in capacitance is due to the fairly long wires, around 8 inches, which I wrapped around one of the screws on the ESD shield so that they wouldn't be accidentally pulled off the chip when the calc is opened.

Does anyone have any suggestions for a way which I can connect the potentiometers to the correct pins on the ASIC, such that:
- there is low capacitance
- if the calc is taken apart, there is no chance of the ASIC being damaged if pulled too hard

here are some pictures of where I put the pots:

Perhaps you could remove the built-in capacitor, and add any additional capacitance needed yourself?

How did you determine excess capacitance due to wires was the problem?

I immediately think of some kind of modular connector. You'd have to fasten the ASIC-side connector securely to something, but I don't think would be too hard. A spring-loaded connector could be very good, since you have the case to ensure correct alignment and a frictionless mate will help avoid damage during disassembly.

A flat flex cable is another thought, but mounting such a thing sounds more difficult if only because FPC sockets are typically surface-mount.

That is an option that crossed my mind, but I hadn't really considered. Do you know what the capacitance of the original capacitor is is, or will I have to figure it out?

Also, I have always thought that the clock was an RC tank, which doesn't make sense with my understanding of tank circuits, you need an inductor involved in addition to a capacitor. What type of oscillator is the calc using?


When I had disconnected both of the potentiometers from the ends of the wires, the problem still persisted. The only way I got it to finally stop was to desolder the wires from the ASIC. I suppose it could also be from inductance in the wires, but capacitance seems like it would have more of an effect here.

Some kind of spring-loaded connector sounds like a good thing, especially with the alignment as you mentioned. I'll look at what I have available.

Some kind of CMOS RC xtal, presumably.