This is an interesting situation to me since back in the 1970's I was
the optocoupler product engineer at Motorola Semiconductor ... now
OnSemi, and of course OnSemi acquired Fairchild several years back.
All LEDs have a finite lifetime. Defects in the crystal lattice
structure propagate (get more plentiful) under stress of current flow
and temperature. The higher the drive current and the higher the
temperature the faster the defects propagate, and those defects cause
carrier recombination instead of conversion to light. The defect
density varies from wafer to wafer and even across the wafer, and is
dependent upon the integrity of the starting wafer and the control of
the epitaxial process that is grown on the wafer. Mechanical damage
during the packaging process (die attach and wirebonding) can also have
a negative effect. ANYTHING that results in lattice defects will affect
the reliability of the LED.
And as mentioned above, using higher drive currents to get more LED
light and therefore more current output on the output phototransistor
will make the device degrade faster.
Reputable semiconductor companies know how to specify and control
lattice damage in LEDs, and cautious users try to minimize drive current
to extend the LED lifetime. I've been out of the semiconductor business
for about 20 years and I'm not saying that ISOCOM and EVERLIGHT aren't
reputable companies ... but I've never heard of them.
All LEDs (defects or not) will have lower light output as the
temperature increases.
The switching speed issue mentioned below is purely a matter of the
amount of light reaching the phototransistor. Less output from the LED
means less carrier generation in the base region of the phototransistor
(i.e., less drive), which of course affects the turn on speed of the
phototransistor. It doesn't make sense to me to worry only about the
two higher speed positions, since if the LEDS are degrading the slower
positions are going to see the same problem soon after the higher speed
positions. The only difference might be if the LEDS in the higher speed
positions are being driven significantly harder.
73,
Dave AB7E
On 6/8/2021 6:42 AM, Kevan Nason wrote:
Thought I’d try to save some people some frustration. Apparently,
optocoupler LED’s have a lifetime. As they get old the light generated by
the internal LED’s weakens and can reach the point where it isn’t strong
enough to cause the receiver to register a pulse.
I purchased a MK2R+ from an internet site that was supposedly in perfectly
working condition. Except it wasn’t. During some very helpful exchanges
with microHAM Customer Service (Jozef particularly), they informed me that
due to a supply issue a batch of MK2R+’s had some alternate optocouplers
installed. Although they had the same specifications as their normal chips,
the alternate parts turned out to have a shorter than expected lifetime.
The unit I purchased would connect to microHAM Router, but after a few
minutes the yellow “X” would show indicating it had lost communications to
the MK2R+. It would automatically reconnect after a few seconds or couple
minutes. The longer I tried to operate the MK2R+ the shorter the times
between dropped comms. That indicated a potential thermally related
problem. After 30 minutes of operation the drop might occur every two to
three minutes. Other symptoms included a continually flashing transmit busy
light (until the MK2R+ reset), N1MM+ would be slow to receive rig frequency
information, inability to transmit, N1MM+ might suddenly change SO2R focus
and/or indicated rig mode, and band switching data fed from the MK2R+ would
be delayed or completely lock on a single antenna despite changing bands.
Also, the internal logs that can be activated from within microHAM router
showed the Mk2R+ was apparently generating an unsolicited PTT signal and
had frequent timeouts during data requests.
Although there are several optocouplers in the unit, only two are used at
high speed and should be replaced. From microHAM:
“Also check what H11L1 optocouplers you have populated in your MK2R+. You
can find two of them in DIP6 package on top board right behind the front
panel POWER button. If they are white than all is OK but if they are black
they can be responsible for issue. White are Fairchild brand, black are
ISOCOM or EVERLIGHT and we found that later deteriorate timing after
several years of use. In this case I recommend you to replace them for
Fairchild brand (now OnSemi but still have F logo on body).”
Mouser had the chips for less than $1 each. It is fairly simple to change
them out. Haven’t used it in a contest yet, but after a three hour static
communication check I had N1MM send CQ into a dummy load for an hour using
a two second repeat. None of the previously noted problems were happening
so it looks like Jozef’s recommendation hit the nail on the head. My unit
is number 478 and made in December 2012. I do not know how long the weaker
chips were used. You can easily see the chips if you pop the cover. They
are on the upper circuit board.
It might be a good idea to proactively check your chips, especially if you
use automatic antenna switching, because this affected how my station
functioned.
Kevan N4XL
_______________________________________________
CQ-Contest mailing list
CQ-Contest@contesting.com
http://lists.contesting.com/mailman/listinfo/cq-contest
_______________________________________________
CQ-Contest mailing list
CQ-Contest@contesting.com
http://lists.contesting.com/mailman/listinfo/cq-contest
|