6551 oscillator

[fridgemagnet]

HI all,

Wondering if someone can assist with a query around the crystal oscillator for the 6551 on a D64. Am fairly sure mine is dodgy, difficult to tell because things seem to spring back into life whenever I attach a probe on it.... Anyway, I've bought a new one unsurprisingly it's somewhat smaller than the fairly chunky one they seem to have fitted back in the early 80's.

However just underneath the existing one appear to be two empty solder pads, which having buzzed them out appear to just link to the pads attached to the larger one. That looks to me like the PCB was designed to take a smaller crystal, should ones become available so to me it looks like I should just be able to de-solder the large chunky one and fit the new small one on the (current) spare pads.

Before posting this question though, I had a quick search on the forum to see if anyone else had done something similar and came across this post from a couple of years back where about half way down the page is a nice picture showing a small oscillator but with the addition of a resistor/capacitor as suggested by a WDC 65c51n document.

Now I'm a bit puzzled, I've got the original MOS 6551 datasheet and it makes no mention of needing additional passive components. Furthermore, measuring the resistance across the existing oscillator shows nothing like a 1M resistance (of course something else could pull that down...). Which then makes me start wondering, is the reason that oscillator is so large in the first place is that includes those extra components inside?

Rgs,

Jon.

[Julian]

The crystal is just a tuned chunk of quartz inside that can, and it just a matter of packaging as to what size is needed and there are variations but fundamentally they are all much the same. An oscillator is a different kettle of fish and should produce a stronger logic level signal, these normally come in 4 pin packages.

Crystals are easily broken with a sharp knock, one of the reasons why they commonly strapped and soldered down to the board. This also means you could potentially trash one by dropping the whole machine - its unlikely and there would probably be other damage but it can happen.

The extra passive components are commonly a pair of small value capacitors that are needed to provide reinforcement of the natural ringing of the crystal. The required capacitance varies but the total passive capacitance is typically about 20pF, without it the crystal may not work reliably or at all. The chip driving it will provide some of that capacitance

[fridgemagnet]

Thanks that helpful. Since there doesn't appear to be any additional cap on the original D64 schematic, sounds like they went with the approach that it'll probably work most of the time and that's good enough. It's also possibly the problem I'm seeing where I think it's right on the edge of sometime working/sometimes not and attaching a probe introduces just enough capacitance to kick it into life all of the time.