Elecraft

KAT100 signal question

Classic

List

Threaded

11 messages Options

BobConvers

KAT100 signal question

In troubleshooting a relay problem on my KAT100, Gary advised me to check the
signal /8R, generated by Q3. I think Q3 may be faulty, since (using a Fluke
DMM) signal /8R alternates between zero and only 4.99v as the KPA100 is
cycled between receive and TUNE. Signal 8R alternates between 7.91v and zero, as
I would expect.

Before, I disassemble, de-solder, and install a new Q3, however, could
someone verify that 4.99 v is indeed lower than it should be and Q3 is probably bad.
Thanks.

73,
Bob, WO3E
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
http://mailman.qth.net/mailman/listinfo/elecraft

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

Don Wilhelm-3

RE: KAT100 signal question

Bob,

I would think Q3 is OK since it simply acts as a switch - it does not
produce any voltage. The MCU (U1) pin 39 is the source for voltage on the
drain of Q3, and a high level of somewhere near 5 volts is to be expected.

Look somewhere else for the problem. It is difficult to see directly
whether the microprocessor is creating the correct output because it is fed
to the relay drivers as serial data, but the relay driver pins should have
the outputs for the selected relays at a low level, and go to about 12 volts
on the pins connected to the unselected relays. Check the schematic sheet 1
to see which pins correspond to which relays.

73,
Don W3FPR

> -----Original Message-----
>
> In troubleshooting a relay problem on my KAT100, Gary advised me
> to check the
> signal /8R, generated by Q3. I think Q3 may be faulty, since
> (using a Fluke
> DMM) signal /8R alternates between zero and only 4.99v as the KPA100 is
> cycled between receive and TUNE. Signal 8R alternates between
> 7.91v and zero, as
> I would expect.
>
> Before, I disassemble, de-solder, and install a new Q3, however, could
> someone verify that 4.99 v is indeed lower than it should be and
> Q3 is probably bad.
> Thanks.
>
>

--
No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.338 / Virus Database: 267.9.4/57 - Release Date: 7/22/2005

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
http://mailman.qth.net/mailman/listinfo/elecraft

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

Jack Brindle

Re: KAT100 signal question

In reply to this post by BobConvers

Q3 is probably not your problem. It is an open-drain switch that
inverts the 8R signal, conditioning it for the microcontroller. It
appears that the MCU provides the pullup for the signal through the
internal pullup on PB6. 4.99 volts is about right for this signal.

Just what is the "relay problem" you are seeing? If Q3 and the /8R
signals were bad, the KAT would not switch to transmit conditions at
all. This would be indicated in a number of ways, such as not even
trying to tune. Relay troubles, however, could be isolated to a small
area in the circuit.

On Jul 23, 2005, at 10:37 AM, [hidden email] wrote:

> In troubleshooting a relay problem on my KAT100, Gary advised me to
> check the
> signal /8R, generated by Q3. I think Q3 may be faulty, since
> (using a Fluke
> DMM) signal /8R alternates between zero and only 4.99v as the
> KPA100 is
> cycled between receive and TUNE. Signal 8R alternates between
> 7.91v and zero, as
> I would expect.
>
> Before, I disassemble, de-solder, and install a new Q3, however, could
> someone verify that 4.99 v is indeed lower than it should be and Q3
> is probably bad.
> Thanks.
>
> 73,
> Bob, WO3E
> _______________________________________________
> Elecraft mailing list
> Post to: [hidden email]
> You must be a subscriber to post to the list.
> Subscriber Info (Addr. Change, sub, unsub etc.):
> http://mailman.qth.net/mailman/listinfo/elecraft
>
> Help: http://mailman.qth.net/subscribers.htm
> Elecraft web page: http://www.elecraft.com
>

- Jack Brindle, W6FB
------------------------------------------------------------------------
---------------------

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
http://mailman.qth.net/mailman/listinfo/elecraft

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

BobConvers

Re: KAT100 signal question

In reply to this post by BobConvers

Thanks, Don. Actually, I get NO action from ANY relay, although the LEDs
are switching correctly. I tried to check the signal RYDATA (as best I could
with no scope, only DMM) and it is an absolutely constant level. The front
panel signal FPDATA jumps around in value as the signals come through to
activate the LEDs, but RYDATA is dead still. There is 12 v on the the hot side of
the relays and 5 volts on pins 2 of U2, U3, and U4. With power removed,
RYDATA measures over 5 megs to ground. Any other suggestions?

73,
Bob, WO3E

>>>>
Bob,

I would think Q3 is OK since it simply acts as a switch - it does not
produce any voltage. The MCU (U1) pin 39 is the source for voltage on the
drain of Q3, and a high level of somewhere near 5 volts is to be expected.

Look somewhere else for the problem. It is difficult to see directly
whether the microprocessor is creating the correct output because it is fed
to the relay drivers as serial data, but the relay driver pins should have
the outputs for the selected relays at a low level, and go to about 12 volts
on the pins connected to the unselected relays. Check the schematic sheet 1
to see which pins correspond to which relays.
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
http://mailman.qth.net/mailman/listinfo/elecraft

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

Don Wilhelm-3

RE: KAT100 signal question

Bob,

I don't have a KAT100 here to check it, but I can say that how much RYDATA
wiggles up and down will depend on what the microprocessor is trying to
initially set the relays to - if it happens to be all '1s' the line may
always be high during the setting time frame. That data is locked into the
shift resisters, so the uP only has to update the settings occasionally. I
would expect the front panel data to be updated more frequently and
monitoring the FPDATA line is much more likely to show activity.

Sorry I can't help more without having it in my hands. Perhaps Gary can
help when he returns to work on Monday.

73,
Don W3FPR

> -----Original Message-----
> From: [hidden email]
> [mailto:[hidden email]]On Behalf Of [hidden email]
> Sent: Saturday, July 23, 2005 2:43 PM
> To: [hidden email]; [hidden email]
> Subject: Re: [Elecraft] KAT100 signal question
>
>
> Thanks, Don. Actually, I get NO action from ANY relay, although
> the LEDs
> are switching correctly. I tried to check the signal RYDATA (as
> best I could
> with no scope, only DMM) and it is an absolutely constant level.
> The front
> panel signal FPDATA jumps around in value as the signals come through to
> activate the LEDs, but RYDATA is dead still. There is 12 v on
> the the hot side of
> the relays and 5 volts on pins 2 of U2, U3, and U4. With power removed,
> RYDATA measures over 5 megs to ground. Any other suggestions?
>
> 73,
> Bob, WO3E
>
>
> >>>>
> Bob,
>
> I would think Q3 is OK since it simply acts as a switch - it does not
> produce any voltage. The MCU (U1) pin 39 is the source for voltage on the
> drain of Q3, and a high level of somewhere near 5 volts is to be expected.
>
> Look somewhere else for the problem. It is difficult to see directly
> whether the microprocessor is creating the correct output because
> it is fed
> to the relay drivers as serial data, but the relay driver pins should have
> the outputs for the selected relays at a low level, and go to
> about 12 volts
> on the pins connected to the unselected relays. Check the
> schematic sheet 1
> to see which pins correspond to which relays.
> _______________________________________________
> Elecraft mailing list
> Post to: [hidden email]
> You must be a subscriber to post to the list.
> Subscriber Info (Addr. Change, sub, unsub etc.):
> http://mailman.qth.net/mailman/listinfo/elecraft
>
> Help: http://mailman.qth.net/subscribers.htm
> Elecraft web page: http://www.elecraft.com
> --
> No virus found in this incoming message.
> Checked by AVG Anti-Virus.
> Version: 7.0.338 / Virus Database: 267.9.4/57 - Release Date: 7/22/2005
>

Jack Brindle

Re: KAT100 signal question

In reply to this post by BobConvers

On Jul 23, 2005, at 11:42 AM, [hidden email] wrote:
> Thanks, Don. Actually, I get NO action from ANY relay, although
> the LEDs
> are switching correctly. I tried to check the signal RYDATA (as
> best I could
> with no scope, only DMM) and it is an absolutely constant level.
> The front
> panel signal FPDATA jumps around in value as the signals come
> through to
> activate the LEDs, but RYDATA is dead still.

It is not surprising that you are seeing the FP serial bus so active
- it is periodically being refreshed. Relays are another matter. The
RY serial bus is rarely (in processing terms) used. It is, in fact,
used only when a relay state needs to be changed. If that doesn't
happen, you won't see any toggles on the RYCK, RYLOAD or RYDATA lines.

> There is 12 v on the the hot side of
> the relays and 5 volts on pins 2 of U2, U3, and U4. With power
> removed,
> RYDATA measures over 5 megs to ground. Any other suggestions?

None of these measurements really say anything. What is the level of
RYDATA, RYLOAD and RYCK at both the MCU and U2, U3 and U4?

Have the relays ever worked, or is this a new unit? You might want to
check the solder joints at pins 15, 16 and 17 of the MCU, as well as
those at pin 3 of U2, and 12 and 13 of U2, U3 and U4. Lastly, cycle
through the K2 menu's C and L settings to check for any relay action.
It is quite possible that the KAT100 is simply not commanding the
relays to change - the question in that case is why not?

- Jack Brindle, W6FB
------------------------------------------------------------------------
---------------------

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
http://mailman.qth.net/mailman/listinfo/elecraft

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

BobConvers

Re: KAT100 signal question

In reply to this post by BobConvers

Jack, as I mentioned to Don, I get NO relay action on ANY relay, as I switch
from ANT1 to ANT2 and /or adjust POWER on the K2/KPA100. The KAT100 front
panel LEDs light up correctly.

I was almost certain that before when I checked I saw no change on RYDATA,
but I checked again and found that RYDATA does indeed switch from steady 5v to
steady zero and back, when I switch from ANT1 to ANT2 and back. POWER adjust
has no effect on RYDATA. FPDATA is at 5v for for all positions of the K2
panel, although FPDATA briefly drops and returns to 5v when I adjust the K2 and
the KAT100 LEDs switch.

>>>>>
Q3 is probably not your problem. It is an open-drain switch that
inverts the 8R signal, conditioning it for the microcontroller. It
appears that the MCU provides the pullup for the signal through the
internal pullup on PB6. 4.99 volts is about right for this signal.

Just what is the "relay problem" you are seeing? If Q3 and the /8R
signals were bad, the KAT would not switch to transmit conditions at
all. This would be indicated in a number of ways, such as not even
trying to tune. Relay troubles, however, could be isolated to a small
area in the circuit.
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
http://mailman.qth.net/mailman/listinfo/elecraft

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

Don Wilhelm-3

RE: KAT100 signal question

Bob,

If you can see a drop on the RYDATA voltage when you switch from ANT1 to
ANT2, you can chase that bit through the first shift resister - look for a
similar drop at U2 pin 18 (and U3 pin 3). It will only drop for the change
from ANT1 to ANT2 (and not the other way 'round), and you get only one
chance to look for it after tapping the button.

The FPDATA lines communicate different data, so there is no proper
comparison between the relay control and the front panel control.

73,
Don W3FPR

> -----Original Message-----
>
> Jack, as I mentioned to Don, I get NO relay action on ANY relay,
> as I switch
> from ANT1 to ANT2 and /or adjust POWER on the K2/KPA100. The
> KAT100 front
> panel LEDs light up correctly.
>
> I was almost certain that before when I checked I saw no change
> on RYDATA,
> but I checked again and found that RYDATA does indeed switch from
> steady 5v to
> steady zero and back, when I switch from ANT1 to ANT2 and back.
> POWER adjust
> has no effect on RYDATA. FPDATA is at 5v for for all positions
> of the K2
> panel, although FPDATA briefly drops and returns to 5v when I
> adjust the K2 and
> the KAT100 LEDs switch.
>
>

BobConvers

Re: KAT100 signal question

In reply to this post by BobConvers

This is a new KAT100, just completed and under test. I have never observed
ANY relay action.
RYLOAD remains at 2.25v and bounces a bit each time I adjust, say, ANT1 to
ANT2.
RYCK appears steady at about zero; and 5Megs to ground with no power.
U3-3 and U4-4 are always zero, also 5Megs to ground with no power.

RYDATA (at both U1-17 and U2-3) is interesting - its action varies according
to setting of the K2 ATU menu. When ATU is set to AUTO, adjustment of ANT1
to ANT2 causes RYDATA to switch between zero and 5v. When ATU is set to C3,
for example, RYDATA remains constant at 5v with adjustment of ANT1 to ANT2
(bounces a bit when switching). However, when I powered down, and back up again,
I could not always reproduce this exact behavior.

After playing around, I reached an ATU menu condition where I could see a
five volt swing as I switched between ANT1 and ANT2. I looked for a similar
swing at U2-18 and U3-3. There was no corresponding swing, always zero. Am I
correct that the finger of suspicion is beginning to point at U2?

>>>>>
None of these measurements really say anything. What is the level of
RYDATA, RYLOAD and RYCK at both the MCU and U2, U3 and U4?

Have the relays ever worked, or is this a new unit? You might want to
check the solder joints at pins 15, 16 and 17 of the MCU, as well as
those at pin 3 of U2, and 12 and 13 of U2, U3 and U4. Lastly, cycle
through the K2 menu's C and L settings to check for any relay action.
It is quite possible that the KAT100 is simply not commanding the
relays to change - the question in that case is why not?
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
http://mailman.qth.net/mailman/listinfo/elecraft

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

Jack Brindle

Re: KAT100 signal question

See notes and questions inside your comments...

On Jul 23, 2005, at 1:10 PM, [hidden email] wrote:

> This is a new KAT100, just completed and under test. I have never
> observed ANY relay action.

What makes you think the box is running at all? What is being
displayed on the front panel, and does the K2 properly recognize the
presence of the KAT100? Are there any bus error messages, or have you
seen any? Also, why do you think the problem is isolated to the
relays and isn't really seen elsewhere?
To put it another way, what _is_ working?

> RYLOAD remains at 2.25v and bounces a bit each time I adjust, say,
> ANT1 to ANT2.

Like RYCK and RYDATA, RYLOAD should tend to sit at 0 or +5V. It is
the slowest changing of the three signals. A voltage of 2.25 volts
would indicate that it is very rapidly changing and your VOM is
averaging it to that level, or something is incorrectly connected on
the board. Again, maybe a bad solder joint, not soldered at all, OR,
a solder splash causing a short from the RYLOAD line to a signal that
is normally high (even +5V). If the RYLOAD signal is stuck at one
level, then the shift registers cannot load the data from the
controller, and would certainly not work.

> RYCK appears steady at about zero; and 5Megs to ground with no power.
> U3-3 and U4-4 are always zero, also 5Megs to ground with no power.

> RYDATA (at both U1-17 and U2-3) is interesting - its action varies
> according to setting of the K2 ATU menu. When ATU is set to AUTO,
> adjustment of ANT1 to ANT2 causes RYDATA to switch between zero and
> 5v. When ATU is set to C3, for example, RYDATA remains constant at
> 5v with adjustment of ANT1 to ANT2 (bounces a bit when switching).
> However, when I powered down, and back up again, I could not always
> reproduce this exact behavior.
>
> After playing around, I reached an ATU menu condition where I could
> see a five volt swing as I switched between ANT1 and ANT2. I
> looked for a similar swing at U2-18 and U3-3. There was no
> corresponding swing, always zero. Am I correct that the finger of
> suspicion is beginning to point at U2?

Not really. The evidence is getting stronger for a short between two
conductors on the board. It is possible that one of the ICs is
internally shorted causing the problem, but it is more likely that
the problem is external to the chips. Go back and inspect the board
with a good magnifying glass, and you might check your solder joints.
Also, check the MCU (U1) to make sure that all pins are in the socket
and that none got bent under when installing the chip. Bending pins
when inserting chips into sockets is way too easy to do, especially
for 40 pin devices. A bent pin could be shorting against another pin,
causing conditions similar to what you described.

- Jack Brindle, W6FB
------------------------------------------------------------------------
---------------------

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
http://mailman.qth.net/mailman/listinfo/elecraft

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

BobConvers

Re: KAT100 signal question

In reply to this post by BobConvers

>>To put it another way, what _is_ working?>>

Sorry, I had not repeated information from an earlier post: I have proceded
in constructing the KAT100 as far as the step of connecting the KAT100 to the
K2/KPA100, on the K2 panel switching back and forth between ANT1 and ANT2,
looking for proper illumination of the tuner LEDs, and listening for operation
of the tuner relays. The K2 menu includes an ATU entry, with proper submenu
entries (AUTO, L1, C1, etc) which I interpreted to mean that the K2 was
recognizing the presence of the tuner. The tuner LEDs correctly indicate operation
of K2 controls (ANT1/2 and HI/tuner relay action from ANT1/ANT2, power adjust,
K2 ATU menu operation, or anything else.

>>
>>A voltage of 2.25 voltswould indicate that it is very rapidly changing and
your VOM is
averaging it to that level, or something is incorrectly connected on
the board. Again, maybe a bad solder joint, not soldered at all, OR,
a solder splash causing a short from the RYLOAD line to a signal that
is normally high (even +5V).
......
The evidence is getting stronger for a short between two
conductors on the board. It is possible that one of the ICs is
internally shorted causing the problem, but it is more likely that
the problem is external to the chips. Go back and inspect the board
with a good magnifying glass, and you might check your solder joints. >>

I had carefully inspected the board with Optivisor, but will check again for
splashes. All solder joints look shiny, but I'll re-heat any joints that are
in the path of signals we are looking at.

>>
>>Also, check the MCU (U1) to make sure that all pins are in the socket
and that none got bent under when installing the chip. Bending pins
when inserting chips into sockets is way too easy to do, especially
for 40 pin devices. A bent pin could be shorting against another pin,
causing conditions similar to what you described.>>

Using the magnifier, the MPU pins look even and properly inserted. I had
considered lifting the MPU out of its socket for a closer look, but thought that
might cause more problems than it would reveal. I'll do it now.

Thanks again, and I will let you know if I find anything.

73,
Bob, WO3E

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
http://mailman.qth.net/mailman/listinfo/elecraft

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com