[Elecraft] K2 #2456 is OFF THE AIR! - Update.

[Dan Barker]

Please don't slap my wrist, but I really couldn't resist (It's the =
weekend!). I didn't see any reason for the K2 not to work with V- and =
I-SENSE removed, and I really wanted to rule out any other problems =
before trying out a new MCU. So, I gently bent pin 5 out from the chip =
and plugged it back in. High current. Damn!

So, I gently bent out pin 4 from the IC (this pin at least goes to the =
hot resistor - I-SENSE) and plugged it back in. Current normal. Cool!=20

I finished up testing and installing the KIO2. Tests out fine. Works =
fine. AT working FB too. In fact, everything is fine except the V and I =
sense. I've had many QSO's with it since. Every few minutes I get the =
beep and the Low Battery warning, but that's nearly all that's funky.

One remaining odd thing is that the RF power output seems to creep up =
slowly with key down. I can't see it on the oscilloscope, but one "store =
bought" RF power meter, one Heathkit power meter, and the RF Probe/DMM =
all show the effect. Could this be a problem? It'll creep up from 4.2 =
watts on key down, to nearly 5 watts 30 seconds later. I'm never key =
down for 30 seconds in operation. My code isn't THAT slow<g>. Is this =
anything to worry about? Is this related to the MCU not having any idea =
what the supply voltage might be?

Interestingly, with IOC pins 4 and 5 connected and the "extra" current =
warming R8 and U3 up nicely, the display current on the K2 matched the =
draw on my power supply nicely. The current sense circuit appeared to be =
working, althoug sucking up 50% of the amount of power needed for the =
whole radio! The display showed v0.08 i0.34. The current display was =
also "correct" with only pin 5 pulled up. (I never could see how pin 5 =
could heat R8 - I guess the two circuits are related inside the chip, or =
failed together, or are linked now after I've melted something<g>).

Curiouser and Curiouser (except support has probably seen it all =
before). In any case, I really don't think there is a short on the =
board. I have done nothing except inspect til my eyes hurt, pull/replace =
the MCU 3 or 4 times, and lift a leg on R8 to rule out a short from or =
through Q11.

Dan / WG4S / K2 #2456-AT-IO.



-----Original Message-----
From: Dan Barker [mailto:dbarker@visioncomm.net]
Sent: Friday, January 25, 2002 11:04 PM
To: Elecraft; support@elecraft.com
Subject: K2 #2456 is OFF THE AIR!


Damn! And I really liked it.

I seem to have let the smoke out of my MCU, and I don't know how or why. =
Makes me think a new one will fail, too.

Support: Please send me U6 for the controller board. Warranty or bucks?

The Group: ANY IDEAS at all will be greatly appreciated (except "bend =
out pin 5 and slam it back in and go ahead on the air sans voltage =
readout").

History:

K2 #2456 went together fairly well. A bit of learning curve on the =
filters, and finally deciding  USB didn't really matter. LSB is fine. CW =
is extremely fine. I even use CW. Every checkpoint resistance, frequency =
and voltage were in spec on the first try. First QSO, 1/18/2002. Receive =
current, 0.22.

So, being lazy, I built the KAT2. Again, all FB right out of the box. =
Only 1 QSO tho, because the KIO2 kit was sitting on the bench. Played =
around with the relay music a bit, but it tuned up all bands on my low, =
40m dipole (well, 3.3:1 on 80m - but that's to be expected). Only QSO, =
1/24/2002. Receive current, 0.22.

On to the KIO2. I wondered about having to cut a land on the control =
board, but the silk-screen clearly stated Revision A, and there was a =
land from Controller board U6/25 to Q5's gate. I'd have thought a 2-week =
old kit would have the latest and greatest boards. The screening says =
1999, tho. Could just be the art.=20

I really don't think I moxied up the board, but it's very hard to see. =
Every nearby land I can check appears fine. However, it powers up with a =
"Low Battery" warning, and pulls 0.34 amp receive. Receiver works fine, =
transmitter works fine, tuner tunes fine, but the Voltage display is =
0.08. Oh yes, the KIO2 doesn't work either. I didn't get into that too =
deeply, because I had that high current to worry about. I'll look over =
the KIO2 with a fine tooth magnifying glass while I scratch my head and =
wait for a chip.

Being concerned about the "mystery" 0.1 amp (that's a lot when you don't =
know where it's going), I inspected the poo out of everything. There =
just is not a [visible] short on the board. Maybe a clipping got under =
something, but I can't figure out where. The inspection shows R8 =
discolored (meter still shows 98.5, so I'll leave it until it fails), =
and U3 somewhat warm.

12v on S1. 12v on U3 pin 5. 0v on U3 6&7. Ov on U6 pin 5. Oh No!

Powered off, U6 pin 5 (V SENSE) shows 2.4 ohms to ground. I know it's =
probably U3 or something on the board, but I pulled U6 anyhow, since =
it's socketed. Now pin 5 (of the socket) shows it's expected 10 Megohms =
(yeah right) to ground. The chip, sitting on the bench in it's foam, =
shows 2.4 ohms from pin 5 to 31 (either polarity on the probed). That =
ain't right!

Can someone give me resistance to ground on all 40 pins of the socket =
with U6 removed (Rev A, land cut, 2 jumpers and a cap)? It's likely not =
that, but I gotta check before putting in another MCU.=20

Very sad WG4S, I guess the Cubic Astro-103 has a few more QSO's in it, =
but I'm really gonna hate twisting all those big knobs on the old =
Griefkit tuner. Gotta put 1/4 inch jacks back on everything too :-(