[time-nuts] My NTGS50AA failed
eb4apl at cembreros.jazztel.es
Tue Nov 11 20:47:31 EST 2014
I meant 2 years, it is quite late here.
I'm not very sure that the DAC is working, I suppose that the unit
doesn't measure the DAC output, it reports the DAC commands. My voltage
figures is what LH reports (so the NTGS50AA reports, probably what it is
trying to do), but the frequency control pin of the oscillator is stuck
at 5.02 V regardless of the supposed (intended) DAC output, it does not
move at all.
I have checked the internal power supply voltages and they are ok.
Hopefully a capacitor could be shorted to the +5V line but it looks too
much luck for me.
Thank you for your suggestions,
On 12/11/2014 a las 2:23, Bob Camp wrote:
> Ok, the gizmo you have is pretty similar to a Thunderbolt in terms of what they did for circuits. It’s by no means identical and I have not traced either one out far enough to have a schematic.
> At least from the TBolt - your DAC is working. The voltage is moving around and it’s getting low enough to bring the oscillator into lock. That suggests that the problem is not the DAC it’s self. I’d bet that what ever magic they do to figure out the oscillator frequency has gone nuts. A very common approach is to use the GPS PPS like a gate on a frequency counter. I think that this part gets a bit more fancy than that. Either way, that sounds like where the problem lies.
> Just in case - always check the regulated power supply voltages on the board and touch test the bypass caps. It’s a silly thing, but it doesn’t take long to do. I know I have gone for a few hours / days / weeks on something like this, only to finally bump into the shorted bypass cap or dead regulator and go “hmmmm…..”.
>> On Nov 11, 2014, at 8:07 PM, EB4APL <eb4apl at cembreros.jazztel.es> wrote:
>> After 2 yeses of continuous operation my NTGS50AA turned its red led on and stopped working. A check with LH shows "OSC: BAD" and "OSC age alarm" assuming that the oscillator had aged too much.
>> I didn't believe this because when working, the DAC control voltage was around 2.9 V, in fact near the middle of its range (0-5 V). Measuring the 10 MHz output it is high, about 4.46 Hz which agrees with the LH figures.
>> I measured the oscillator EFC pin and it is struck at 5.02 V, not following the DAC voltages as reported by LH. I think that either the DAC is bad or an amplifier after it. Since I don't have any schematic and the oscillator covers the top layer the troubleshooting is difficult.
>> Has anybody experienced this failure before?. Does anybody has an schematic, even a partial one?
>> Since the unit now is operating open loop I observed the locking strategy of this GPSDO. First it waits about 12 minutes for warming (I don't know if it internally monitors the oven current or uses a fixed time). During this period it sets the DAC output to the initial value as stored in the EEPROM (3.0 V).
>> When it thinks that it is "warmed enough", the DAC is ramped in the right direction to intercept exactly 10 MHz (towards 0 in my case). If it reaches 0 then declares the alarm and the DAC voltage is set to half the initial value (1.5 V). Five minutes later it switch the DAC to 2,25 V and 17 seconds later it returns to 1.5 V and remains there for 4 min and a half. Then it goes up again to 2.5 V and after 10 seconds it goes down to 1.88 V and after some 14 minutes it goes down again to 1.5 V, it remains there for about 15 minutes and then it goes down to 1.14 V, remains there for 80 seconds and goes back to 2.5 V.
>> It looks like it checks from time to time if it is able to control the oscillator or simply it does weird things once it thinks that the oscillator cannot be disciplined.
>> I will appreciate very much any information.that can help my troubleshooting. These units has more than doubled its price since I bought mine and I think that they are vanishing. The Guatemala's cell towers scrap has been exhausted.
>> Ignacio EB4APL
>> 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.
More information about the time-nuts