[Elecraft] [K3] Disappearing VCO error

[Frank R. Oppedijk]

Hi group,

For some months now, I had the same intermittent problems as Ian: after the K3 has been running for a few hours, it stops working with ERR PL1 (and PL2). The next morning, after cooling off, things are back to normal. Up to yesterday, I was always able to make this issue go away instantly by opening up the K3 and giving all cables going to the KREF3 board a twist. But not yesterday...

Thanks to Ians email, I tried removing the KREF3 board, and removing and reseating the oscillator. I also gave all connectors (they were a bit oxidized) a very light scrub with fine sandpaper and then wiped them with a cloth with some anti-corrosive spray. After building things together again and doing the synthesizer re-calibration, my K3 now is back in working order.

73 de Frank PA4N


On 28 okt. 2010, at 14:41, Ian White GM3SEK wrote:

> Yesterday morning, I wrote:
>> 
>> After the K3 had been running for a few hours yesterday evening, just
>> listening to the bands, it stopped working with ERR PL1 (and PL2). These
>> were genuine hardware errors - neither of the two VFOs would tune and
>> both receivers were dead. Attempts to re-calibrate the VCOs ended with
>> ERR VCO and ERR VC4.
>> 
>> I switched off and left the problem overnight, and this morning both
>> VCOs calibrated successfully with no further errors so far.  (A notable
>> difference was the successful calibrations took much longer over some of
>> the steps. The failed attempts had rattled through all the steps at the
>> same high speed.)
>> 
>> The firmware was MCU 4.12 / DSP 2.60 / FPF 1.13.  After everything was
>> working correctly again, I have updated to MCU 4.14.
>> 
>> Hardware is K3/100 #724, with KAT3, KRX3, KXV3 and all modifications to
>> date.
>> 
>> The question is: why did this happen? And with CQWW SSB coming up, could
>> it happen again?
>> 
> 
> By 10am Aptos time, answers from Elecraft were in my inbox!
> 
> Dale confirmed that the faults as described were almost certainly due to 
> loss of signal from the 48MHz reference oscillator on the KREF3 board. 
> Since I had been doing some unrelated mods that required the removal and 
> replacement of the KRX3 (several times over) it is very possible that 
> the oscillator module had been knocked, and that contact in the socket 
> had become marginal.
> 
> Eric added that the Clock to Data timing on the internal serial bus is 
> critical to proper operation, and that recent firmware releases include 
> an additional timing margin.
> 
> On opening up the K3 again, there wasn't any obvious problem with the 
> seating of the oscillator module, but I removed and re-inserted it 
> anyway. Instead of the single tie-wrap that was provided in the kit to 
> secure the module in its socket, I tried using two slightly thinner ones 
> which ensure that the module is pulled firmly downward at *both* ends of 
> the package.
> 
> Grateful thanks to Dale and Eric for such a rapid and accurate response!
> 
> Cross fingers for the weekend... and good luck in the contest to all K3 
> users.