[time-nuts] Introduction of Z38XX

Ulrich Bangert df6jb at ulrich-bangert.de
Thu Jun 25 02:52:29 EDT 2009


in the meantime I have received some feedback. Please read it with my
answers in case you noticed something similar:

> 1. The vertical scale has some issues. It goes down from 10-8 
> to 10-23 and then goes up again to 10-7.

Read the 10-23 as "near zero". Everything below should have a "-" sign,
indicating negative values. I enclose a picture how it looks here. 

> The EFC value is remains fixed at 
> 65000. My Z3801 goes to EFC 696364 so stays at the top always.

All scales are "autoscaling" and should follow whatever values you have. You
mave have however put the autoscaling out of work by zooming into the
grahics or by panning the graphics. Zooming is initiated by pressing the
left mousebutton at the top left point of the area to be zoomed and then
draw a white rectangle to the bottom right point of the area to be zoomed.
Panning is initiated by pressing the right mouse button within the graphics
and then move the mouse while the button is pressed. You can use both
actions as often as you want and return to your initial screen by "reverse
zoom" action, i.e. press the left mouse button anywhere in the graphics and
move the mouse left and up while the button is pressed. A second possibility
were that you have played around with the chart editor and switched the
scale to a fixed value. In this case you can turn everything to the start
values by termintaing the program and then start it again. All existing data
will be read in again so will only loose one value or two

> 2. The holdover unc. and sats the sat scale is not OK. It 
> seems to indicate twice the number of tracked sats and a high 
> number (76) of visible sats.

First check whether the visible and tracked number of sats is shown correct
in the main screen's status bar. If the error is already there then check in
the parameters whether the "Where to look for data in the status screen"
values are correct for you (You can count the lines in the receiver status
screen). May be that I initialized these values wrong as long as no
Z3801.Ini is available.

73 de Ulrich, DF6JB   

> -----Ursprüngliche Nachricht-----
> Von: time-nuts-bounces at febo.com 
> [mailto:time-nuts-bounces at febo.com] Im Auftrag von Ulrich Bangert
> Gesendet: Dienstag, 23. Juni 2009 11:29
> An: Time nuts
> Betreff: [time-nuts] Introduction of Z38XX
> Gents,
> perhaps you have read that since a few days I am the owner of 
> a Z3805 bought on eBAy from fluke.I. 
> The other day a friend of mine visited me and brought his 
> GPSCON with him to test it on my Z3805. As fluke.I had 
> pointed out, the GPSCON would work with the Z3805 but it 
> would not work really satisfying. The reason is that the 
> Z3801 has a 6 channel receiver and the Z3816 has a 8 channel 
> receiver, so GPSCON is simply not prepared to read more than 
> 8 sat data sets from the Z3805's status screen (which has a 
> 12 channel receiver). In addition, because the Z3805's status 
> screen uses more lines for the display of the additional sat 
> data sets GPSCON will miss the last three lines of the 
> z3805's status screen.
> This in conjunction with some other things encouraged me me 
> to write a z3805 tool of my own named Z38XX. Of course it 
> will handle z3801s and z3816s as well. It is far from being 
> complete but you can get a good impression of it by 
> downloading it from the usual place
> http://www.ulrich-bangert.de/html/downloads.html
> Z38XX talks to the Z3801/5/16 with a exact 10 s time base 
> which means the PPS TI delivered can be used for the 
> computation of Allan deviation (and other measures) of the 
> internal phase comparator with a Tau0 = 10 s (which is done 
> internally in Z38XX similar as in Lady Heather).
> The sat's horizont line in the Azimuth / Elevation chart will 
> look strange in the first hours after program start. Give it 
> some days until it looks as attched or even better.
> Have fun
> Ulrich Bangert
> www.ulrich-bangert.de
> Ortholzer Weg 1
> 27243 Gross Ippener 

More information about the time-nuts mailing list