Electronics > Projects, Designs, and Technical Stuff

Abused FE5680A

(1/2) > >>

grumpydoc:
My FE5680A is feeling a bit abused. I'd made up a small interface board with a 7815 & 7805 regulator, brought out the 10MHz to a BNC etc and it had been perfectly OK for ages.

Then it ran into a bit of bother. First the 5V regulator died shorted and fed.... well a lot more than 5V to that rail for at least 12 hours overnight before I noticed the "power on" LED was glowing much too brightly...

Amazingly it seemed to survive that once I'd replaced the regulator.

Then a spider crawled into the fan and died messily gunging the fan to a halt with what was left of its body jammed between one of the fan blades and the side wall of the fan (largish UK garden spider). Again it did this overnight so I didn't notice the unit was without any cooling for at least 12 hours - it is bolted to a heatsink though so I wasn't overly concerned but it was quite toasty when found it the next day.

Then to add the final insult the 5V died again, this time because the ground lead fell off the regulator - not sure whether that caused any overvoltage on the 5V rail or just caused it to run with only the 15V supply for a while.

Again I was slightly surprised and pleased to discover it still works. Especially as they've gone up in price again - when Dave did his review/teardown they were $50, by the time I got mine they were £50 and they're now at least £100 in the UK (still from the Chinese vendors on ebay).

Well, it nearly works - instead of the lock output swinging nicely from about 4.7V to 0V it goes from 2.9V high to 1.2V low, so no longer drives the transistor/LED it's connected to correctl and the "lock" LED is on all the time.

Looks like I'm going to have to open it up.

Has anyone traced this part of the circuit out in more detail that the partial schematic here http://www.rhodiatoce.com/pics/time-nuts/FE-5680A/FE-5680A_schematics_v0.3.pdf, or aware of a more complete schematic. Looking at the circuit I'd expect that the PPS signal will be dead as well but I don't use that.

bingo600:
I connected a FE5680A to a 15v switcher (laptop like) psu , that i got from a surplus place.
What i didn't test was the polarity on the cable   |O

That stupid PSU designer put + on the braid , and - on the core  :--

My FE5680A didn't like that and has now departed for the "Silicon heaven"

That'll teach me not to assume , but measure :palm:

I'm glad i got a few more when price was 40$

/Bingo

Cones:
Hi,

I am on my third unit. The first two did not work. The third one does not give 10MHz but 8.66MHz something or other and I cannot get the interface working on RS232 to change it.

I might have to have a go at troubleshooting the first two. They both light up the Rubidium lamp but refuse to lock.

All three are different internally.

Mark

grumpydoc:
Ouch!

Thankfully mine seems a simple fix - there's a 74ACT240 which is connected to the lock signal as a buffer before that signal is (also) used to gate the PPS output. That input was sitting at about 2V and pulling the lock line up/down. For the time being I've just lifted the pin while I find a replacement - it's working happily now. For some reason I've loads of 244's sitting on various boards, but no 240's.

grumpydoc:

--- Quote ---They both light up the Rubidium lamp but refuse to lock.
--- End quote ---

Do they produce an output at about 10MHz?

Apparently things can go out of tune so that the VCO doesn't sweep through 10MHZ and then they won't lock but this is fixable.

Loads of very useful info and links here - http://www.ko4bb.com/dokuwiki/doku.php?id=precision_timing:fe5680a_faq if  you haven't already found that page.

Navigation

[0] Message Index

[#] Next page