[time-nuts] FE-5680B Rubidium and DDS

Clint Jay cjaysharp at gmail.com
Wed Sep 30 16:06:29 EDT 2015


The 9851 will work with pretty much any clock and give an output but of
course you need a decent speed clock to get a useful range of frequencies.

The handy feature of the 9851 is that it has a clock multiplier built in so
that 30MHz can clock it at its maximum speed and give a useful DC-60MHz
output.

I'm a little past the point where the author is, I'm working my way through
the CPLD dump..

On 30 September 2015 at 20:54, Bryan _ <bpl521 at outlook.com> wrote:

> Clint, Yes, same seller. Yes schematic is partial, but it did answer a
> question for me that I posted on eevblog where others suggested removing a
> number of components to feed the 20Mhz to a divider. Looks like its a
> resistor, cap and inductor from the schematic; R1,C1,L1.
> I see, you want to make a easily reprogrammable frequency generator. I
> haven't looked at the datasheet of  the AD9851, but thought it needed a
> 125Mhz crystal. Probably just easier to use a circuit like this.
> http://www.perdrix.co.uk/FrequencyDivider/index.html
>
> http://www.ka7oei.com/FE-5680_diag_1c.gif   (would still need to divide
> first)
> I was planning on the same, but just using the /2 for a 10Mhz out in the
> first link.
> Was always hoping for a easy mod inside the "B", but from the write up
> that was just posted, the author has scoped it pretty well and it appears
> the disappearing 10Mhz appears to be firmware driven.
> Cheers
>
> -=Bryan=-
>
> > Date: Wed, 30 Sep 2015 11:45:49 +0100
> > From: cjaysharp at gmail.com
> > To: time-nuts at febo.com
> > Subject: Re: [time-nuts] FE-5680B Rubidium and DDS
> >
> > Bryan, I think we do, bought from an eBay seller named Fluke.I?
> >
> > I also found that schematic but it's a partial, very useful nonetheless.
> >
> > There's also a dump of the firmware and CPLD so I'm hoping I can make
> some
> > sense of that to work out the CPLD, I'd be happy with a solid 10MHz
> output
> > but to my mind, if I can clock an AD9851 with the 30MHz signal I can get
> > 'as good as' with the same or better performance to the serially
> > programmable DDS module.
> >
> > My thinking may of course be flawed, I'm new to the whole time nut game
> and
> > am learning as I go.
> >
> > The M09 chip seems to be in place of the MAX3232 chip, coupling RX/TX
> from
> > connectors to the CPU, it may be an opto but the package is tiny so it'd
> > not offer much isolation, I'll take another look this evening and work
> out
> > the connections fully.
> >
> > On 29 September 2015 at 21:06, Bryan _ <bpl521 at outlook.com> wrote:
> >
> > >
> > >
> > >
> > > Hi Clint:
> > > Looks like we have the identical board. I suspect the XC9572 is turning
> > > off the 10Mhz, although why would it wait for 6 seconds?. And as you
> say
> > > it's not related to the "lock" . Here is a schematic I found for a
> FE-5680A.
> > >
> > >
> http://www.rhodiatoce.com/pics/time-nuts/FE-5680A/FE-5680A_schematics_v0.1.pdf
> > > I also note the same strange chip in the middle of the absent Max233,
> > > wonder if it is some type of optoisolator to protect
> > > inputs/outputs?Although not sure why one would be needed if the MAX232
> is
> > > unpopulated.
> > >
> > > -=Bryan=-
> > >
> > > > Date: Tue, 29 Sep 2015 15:41:37 +0100
> > > > From: cjaysharp at gmail.com
> > > > To: time-nuts at febo.com
> > > > Subject: Re: [time-nuts] FE-5680B Rubidium and DDS
> > > >
> > > > I've made some investigations and re-confirmed my earlier findings
> on the
> > > > FE-5680B I have.
> > > >
> > > > It locks from cold in under five minutes and runs at approximately
> 700mA
> > > > steady with the expected PPS signal on the output connector when
> locked.
> > > >
> > > > The particular 5680 I have has a Hi Density 15p male connector on the
> > > > output and uses a single 15V supply, part number FE5680B UN 77672.
> > > >
> > > > The device does not have the optional DDS daughterboard which means
> that
> > > > it's not usable as a programmable signal source as it stands (but see
> > > > later).
> > > >
> > > > For approximately six seconds after power is applied a 10MHz signal
> > > > available on the output connector, this signal comes directly from
> the
> > > > XC9572 CPLD via a filter network and appears to be a clean sine wave.
> > > >
> > > > The 10MHz signal seems to be 'switched off' by some event in the
> unit, I
> > > > don't yet know if this is in response to an event or just a timed
> > > shutdown
> > > > but power cycling the unit re-enables the 10MHz output for a further
> six
> > > > seconds. It doesn't appear to be related to rubidium lock as it takes
> > > > consdierably longer than six seconds to lock.
> > > >
> > > > There is a solid 30MHz signal inside the unit, stable once the
> rubidium
> > > has
> > > > locked and the PPS output is also available once locked.
> > > >
> > > > Other observations show that the MAX3232 serial driver chip is
> missing
> > > and
> > > > in it's place there is a tiny 8 pin device marked M09 or MO9 which is
> > > > connected to the TXD/RXD pins on the output connector and the TXD/RXD
> > > pins
> > > > on the 80C323 CPU.
> > > >
> > > > I'd be very interested in suggestions as to the part number of that
> > > device,
> > > > it may yield clues as to the communication method needed.
> > > >
> > > > Attempts to communicate with it via a terminal program have given no
> > > > responses at several 'standard' baud rates. Voltage levels are LVTTL.
> > > >
> > > > I'm wary of applying 5V to any of the pins on the interface connector
> > > > directly and probing them with 3.3v via a 10K resistor has made no
> > > > difference to any of the signals I can monitor (30MHz, PPS etc.)
> with the
> > > > exception of pin 13 which I *think* is reset, obviously this causes
> the
> > > > frequencies to skew for a few seconds until reset is completed.
> > > >
> > > > My intent is to use the 30MHz signal from the CPLD to clock a DDS
> chip
> > > > (probably one of the eBay DDS modules) that's controlled by a PIC
> chip (I
> > > > already have code to run a DDS VFO I developed earlier this and late
> last
> > > > year. Hopefully this will be small enough to fit inside the casing
> > > though I
> > > > don't see a problem with bringing the 30MHz signal out if necessary.
> > > >
> > > > My apologies if this is outside of the scope of this list, I will
> also be
> > > > writing up my findings and experiments with this standard on my blog
> if
> > > > anyone is interested?
> > > >
> > > >
> > > >
> > > > On 26 September 2015 at 11:23, Bryan _ <bpl521 at outlook.com> wrote:
> > > >
> > > > > Thanks Hal, that's interesting, will try and see how far I can get.
> > > > > -=Bryan=-
> > > > >
> > > > > > To: time-nuts at febo.com
> > > > > > From: hmurray at megapathdsl.net
> > > > > > Date: Fri, 25 Sep 2015 23:40:40 -0700
> > > > > > CC: hmurray at megapathdsl.net
> > > > > > Subject: Re: [time-nuts] FE-5680B Rubidium and DDS
> > > > > >
> > > > > >
> > > > > > > I as well wish there was a quick way of converting it back to
> > > 10Mhz. I
> > > > > am
> > > > > > > sure it can be done, just not sure how or where to look
> > > > > >
> > > > > > If the 10 MHz is visible for a second or two, there is probably a
> > > gate to
> > > > > > turn it on/off.  If I wanted 10 MHz, I'd open it up and trace the
> > > wire
> > > > > back.
> > > > > > If it goes to a gate, you might be able to lift the pin for the
> other
> > > > > input
> > > > > > and wire it hi/low.  There is a good chance it goes to a FPGA
> where
> > > you
> > > > > can't
> > > > > > get at the gate.
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > > --
> > > > > > These are my opinions.  I hate spam.
> > > > > >
> > > > > >
> > > > > >
> > > > > > _______________________________________________
> > > > > > time-nuts mailing list -- time-nuts at febo.com
> > > > > > To unsubscribe, go to
> > > > > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> > > > > > and follow the instructions there.
> > > > >
> > > > > _______________________________________________
> > > > > time-nuts mailing list -- time-nuts at febo.com
> > > > > To unsubscribe, go to
> > > > > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> > > > > and follow the instructions there.
> > > > >
> > > >
> > > >
> > > >
> > > > --
> > > > Clint.
> > > >
> > > > *No trees were harmed in the sending of this mail. However, a large
> > > number
> > > > of electrons were greatly inconvenienced.*
> > > > _______________________________________________
> > > > time-nuts mailing list -- time-nuts at febo.com
> > > > To unsubscribe, go to
> > > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> > > > and follow the instructions there.
> > >
> > >
> > > _______________________________________________
> > > time-nuts mailing list -- time-nuts at febo.com
> > > To unsubscribe, go to
> > > https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> > > and follow the instructions there.
> > >
> >
> >
> >
> > --
> > Clint.
> >
> > *No trees were harmed in the sending of this mail. However, a large
> number
> > of electrons were greatly inconvenienced.*
> > _______________________________________________
> > time-nuts mailing list -- time-nuts at febo.com
> > To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> > and follow the instructions there.
>
> _______________________________________________
> time-nuts mailing list -- time-nuts at febo.com
> To unsubscribe, go to
> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts
> and follow the instructions there.
>



-- 
Clint.

*No trees were harmed in the sending of this mail. However, a large number
of electrons were greatly inconvenienced.*


More information about the time-nuts mailing list