[time-nuts] FEI5680A Rubidiums On Ebay
J.Bastemeijer at EWI.TUDelft.nl
Fri Sep 16 05:57:37 EDT 2005
I just looked up the AD9832. It has no internal multiplier. Maximum
reference CLK-frequency is 25 MHz. Squeezing out 10 MHz...... the wave
shape will not be nice, unless extensive filtering is used.
I'm not sure if this DDS is used at the end of the "chain" or is a part
of the frequency controlloop? If it is at the end of the chain, one
could consider disconnnecting the program-pins and program a small PIC
to "program" the DDS.
Or use the reference clock and connect it to a more recent type of DDS
(with internal multiplier), and create a nice 10 MHz sine-wave with the
I have seen these standards on Ebay as well some time ago.... I expected
all these kind of problems, so I decided not to buy one. However, if the
price would have been lower (and if I had more time....), I would have
bought one. My idea was to modify the unit. Keep some of the original
electronics, but redesign the synthesizer and interface part....
Good luck, hope you will find some way to make the oscillator usefull!
Best regards & 73's Jeroen - PE1RGE
>On Thu, 15 Sep 2005 16:21:51 +0200, Jeroen Bastemeijer
><J.Bastemeijer at EWI.TUDelft.nl> wrote:
>>By the way, did someone look into the type-number of the DDS? Or is it a
>>custom build "discrete" DDS?
>The FEI5680A uses an AD9832. There is also a Xilinx in it and a couple
>other big ICs. The output on the one I bought is 1 PPS and 10 MHz. The
>FEI web page specs a frequency of 1 to 20 MHz and programmable to 1 x
>10^-13. FEI has not been friendly in giving any more info.
>The eBay seller claimed that the unit needed 15V and that info on the
>FEI web pages described the connector. After returning one, I did a some
>reverse engineering and figured out that this unit needed 15V *and 5V*
>to run. I think I have figured out the pins on the D connector, and it
>does not exactly match the info on the FEI site.
>Seller claims other satisfied buyers, but I can't imagine how they
>figured out how to use it unless they have access to some Motorola
>documents. (It seems it was use in Motorola equipment.)
>There is a MAX3232 serial chip in it that connects to the D connector.
>The seller claimed originally that they programmed like the 5650A, but
>this is not true. Nothing I ever sent to it on the serial lines ever
>elicited any response or reaction from the unit.
>Mine is not exactly on frequency (10 MHz) so it would be nice to have a
>way to adjust it.
>Here is part of a message I posted after my crusade to figure this thing
>out in January...
>To get to the point... by digging at circuits, I eventually figured out
>they needed +15 V but also +5 V on a different pin. With the two
>voltages, mine now came up giving 1 PPS and also 10 MHz on a different
>I have worked out that RS232 Tx and Rx signals (from a MAX3232) come out
>to two pins on the external DB9 connector. I had expected it would
>program like the 5650 as described here:
>Thus far, I haven't seen any reply to anything I send to the unit on the
>So, either these units are not programmable or I just don't know how to
>talk to them. I have not found any documentation on how to program the
>5680A. Does anyone have access to any more information. Did Don Latham
>work out his published information about the 5650 by logic and guess
>work or did he have access to something more?
>I wanted to try all other sources of information before contacting FEI
>as I have heard they are not that friendly unless money is involved.
>If it helps, this FE 5680A rubidium was used on a Motorola SGLA4000B
>High Stability Oscillator (HSO) that was (I think) used in cell sights.
>time-nuts mailing list
>time-nuts at febo.com
Ing. Jeroen Bastemeijer
Delft University of Technology
Department of Electrical Engineering
Electronic Instrumentation Laboratory
Mekelweg 4, Room 13.090
2628 CD Delft
E-mail: J.Bastemeijer at EWI.TUDelft.nl
More information about the time-nuts