Moderators: Dave Moll, Andrew Davie, Steve Anderson
Steve Anderson wrote:I think you've got that around the wrong way, the Philips/NXP versions are better. But even so, in this application both are fine, we don't need anywhere near 10MHz.
Steve A.
Steve Anderson wrote:It has to be a HC4046 version (or HCT4046), as does all the other logic, CD/HEF simply won't work. The maximum frequency you'll get out of a CD/HEF version is 1MHz at 5V, maybe 2,5 at 12V. We need around 3.5MHz at 5V.
Steve A.
Steve Anderson wrote:Harry, first part of the timebase logic - the base frequency generation. You'll more than likely have to fiddle around with the values of R101, R102 and C101. The cap having the biggest influence. This should give you a frequency as noted with the pot in the centre of its track and a reasonable variation plus and minus. Do not take R101/102 below 3k0.
Keep stray capacitance, especially around C101 to an absolute minimum. If you're using a socket for IC101 I'd be tempted to bend pins 6 & 7 upwards and 'tack' the cap onto them.
Steve A.
ppppenguin wrote:If your HC4046 is struggling you can tweak the supply voltage. Run your logic at 5.5V and it will all go that little bit faster.
I did something similar with a DAC once, in a professional design. It potentially needed to run at 81MHz but at 3.3V it was only spec'ed to something like 75MHz. If it was a one off you just try it and see if it works but this was for series production. You could run it up to 5V but then power disspation went up and compatibility with the 3.3V logic driving it was marginal. So I ran it at 3.8V. Good for 81MHz according to the data sheet but still OK for interfacing to 3.3V logic.
Steve Anderson wrote:As long as they are the HC version, they'll be OK for this. The CD/HEF versions won't be. The company I use for chips doesn't stock CD/HEF any more, special order only, minimum quantity 1000, delivery one month. You can tell they just don't want to carry the stuff as it's slow moving and ties up capital...and stock room space.
HC will probably go the same way in the not too distant future.
Steve A.
Steve Anderson wrote:Harry, here's the second part of the timebase circuits. Be sure not to mix up the supply voltages. I've coloured them hoping that it will stand out. Don't forget good decoupling/bypassing on all three supplies.
Keep the lead-length to/from SW101 as short as possible.
I've just noticed an error, the three lines going to SW102 should be 1I, 1J and 1K, the frequencies are correct. I'll update the master AudoCad file.
Steve A.
Return to The Tasmanian Devil VCR139....A for Andrew
Users browsing this forum: No registered users and 1 guest