Hi I would be very careful with the reference conversion.
Your rubidium probably puts out 10 MHz and the counter needs 1 MC .... The 4000 series CMOS gates probably have quite a bit more jitter than some of the more modern parts. Stuff like the 74AC series will beat them on jitter. Bob On Jan 10, 2010, at 11:48 AM, paul swed wrote: > Arie > Still working on the a26 board. It works but I want to improve the 10 mc in. > Most likely a schmitt trigger CD 4093 I might guess. The boards very simple. > Since I have 10 mc in the shack from a RB standard I simply square it up and > send it to a CD4050 buffer IC to the 10 MC systems in the 5248M. Then take > the same 10 MC divide with a CD 4018 to get 1 MC feed this to 3 CD4050 > inverters. I strongly suspect actually one buffer is more then enough in the > package just have not had time to try it. > > I also found some HP sloppy work. HP no way. But evidently it does happen. > There is a 7474 on the A35 board that has open set and reset inputs. Good > practice is to tie those high. > These are floating and as I had seen with age they become slightly erratic. > Replaced that chip and will tie the spare leads high. > > I have indeed isolated my issue to the tunnel diode gate section. Inject a > signal just beyond that and the counter counts. Though incorrectly because > no gate action. > > I have another 5248 thats been working for years and had never been in it. > Well I figured I would borrow the A21 card to confirm the issues and measure > voltages. Son of a gone someone else had the same section issue and cleverly > replaced the tunnel diode with a ecl 10116. I don't see how they are gating > that yet. But I do understand how the 10116 works to amplify and clip the > signal. So I may indeed copy that circuit if I can not recover the gate > section of A21. > Regards > > On Sun, Jan 10, 2010 at 11:38 AM, paul swed <[email protected]> wrote: > >> Arie, >> >> >> On Sun, Jan 10, 2010 at 7:14 AM, arie schellaars < >> [email protected]> wrote: >> >>> >>> >>> >>> >>> Hi folks, >>> My 5245L had faulty nixies and my approach was to use Russian replacements >>> available from a Canadian outfit. >>> The "snags" are. >>> These nixies , though a direct socket fit , have a glass extrusion at the >>> back to >>> "suck out" or let in the neon gas during manufacture. >>> I drilled the socket ,slow and carefully here, to allow this extrusion to >>> fit into the >>> socket . There is some room to spare for this on the socket and no >>> disassembly of the readout card is needed. >>> The second "snag " is that the figure 5 is in fact an inverted (upside >>> down) fig 2 but >>> in practice it's not a problem. >>> BTW never had a problem with any of the "innards" of the counter boards! >>> Would be interested in your details of the oscillator board as , like you, >>> mine came without the crystal. The oven container was there but empty! >>> Have since obtained a crystal in its oven but still not happy with the A26 >>> board. >>> performance AND have run out of Germanium transistors as per on that >>> board. >>> >>> Just had a thought. Ever since I purchased my 5245L many years ago, I have >>> been on this Time nut treadmill . I guess I should have known better at the >>> time !!! >>> Now with a shack full of HP and Tek gear and the inevitable rubidiums , >>> GPS thunderbolts and plain old oxco's , time to call "CQ' is at a premium. >>> (Hi) >>> Cheers from OZ >>> Arie >>> VK3DBF >>> >>> >>> Been troubleshooting a 5248 M someone grabbed teh oscilator and A26 board >>> from it. >>> The a26 board was replced with 2 cmos chips and 1 transistor. >>> Now I am at the counter gate. Darned if I don't have an issue in the >>> dreaded >>> tunnel diodes area. Such a lucky day for me. >>> Need to cleanup and solder in the pseudo a26 board so I can more easily >>> work >>> on the gate. >>> >>> >>> >>> >>> >>> __________________________________________________________________________________ >>> See what's on at the movies in your area. Find out now: >>> http://au.movies.yahoo.com/session-times/ >>> _______________________________________________ >>> time-nuts mailing list -- [email protected] >>> To unsubscribe, go to >>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >>> and follow the instructions there. >>> >> >> > _______________________________________________ > time-nuts mailing list -- [email protected] > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. > _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
