[time-nuts] Minor TADD-2 Sync Problems
rdarlington at gmail.com
Wed Sep 16 21:55:54 UTC 2009
These times are 400ns apart which corresponds to the duration of the
assembly instructions running at 10MHz (or 800ns which corresponsd to the
duration at 5MHz) on a PIC microcontroller. TVB has a blurb about this
somewhere in his PPSDIV assembly code which the TADD-2 is based on.
On Wed, Sep 16, 2009 at 3:48 PM, Brian Kirby <kilodelta4foxmike at gmail.com>wrote:
> I have built two TADD-2's and have been using them in the timing shop. I
> noticed a synchronization problem and I wanted to pass it on and see if
> anybody else seen the same problems.
> What I noticed is the units seem to want to sync within three discreet
> values of time ; of the source. I have to press the sync button and watch
> the output to confirm where its set. I have to press the sync button
> several times to get the proper value. Attached is a drawing of the test
> The data is for two TADD-2 units, I will call them units A & B.
> Using 10 Mhz input , unit A will either sync to 171 nS, 571 ns or 971 ns.
> Unit B will sync to 71 nS, 471 nS or 871 ns.
> Using 5 Mhz input, unit A will sync 400 ns, 1.200 uS or 2.000 uS. Unit B
> will sync to 600 nS, 1.400 uS or 2.200 uS.
> The Austron 1210 shows 3.5 volts P-P terminated sine waves on the scope for
> 10 and 5 Mhz, the 1 PPS is +5.8 volts peak, also terminated.
> Both TADD-2 were acquired and built less than a month ago - so the parts
> should be identical values. The only difference is power supply. Unit A
> runs off my 12 volt system that powers my GPS receivers. Unit B has a 9
> volt DC wall wart rated at 0.5A.
> I changed the RF Input to my HP5065A and I still see three different values
> of time; Unit A -102 nS, 698 nS, and 1498 nS. Unit B shows 502 nS, 1302 nS
> and 2102 nS.
> I change to other RF and PPS sources and I see the same problem.
> My concern is this is repeatable, in my case, and I have to watch the
> output to get the sync as close to the source, timing wise with the TADD-2.
> My other dividers, based on async logic do not show this problem. One
> unit syncs solid at 160 nS on 10 Mhz, the other unit at 210 nS (they were
> built several years apart and probably different brands of gates).
> Brian - KD4FM
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to
> and follow the instructions there.
More information about the time-nuts