Damaged K3S

[K4JST]

Looking for resources for trouble shooting my K3S.

During FD this year my K3S which was being used as the CW station for WAARC/K4RC stopped working and emitted blue smoke!  Autropsy revealed a cracked D28 (one of the diodes to protect against reversed polarity and transients).  Contacted Elecraft and they had none, but suggested Mouser and Digikey.  Mouser listed the diode with a 45 week delivery tme!! Digikey had 18 and I bought two.  Replaced the diode and the rig came up, however, it will not turn off (needs to be unplugged).  Also, Though the innitial display show 40 cw, there is no audio and if I try to go to the menu, display goes blank.  Tried a master reset (shift/LO followed by Power) I do get EE INIT, but no error codes. EE INIT immediately dissapears and is replaced by HI RFI (no matter how quickly I release the POWER button) from thjis point on, nothing seems to work.

 

Does anyone have a suggestion for what I should look at next?  Is there an Elecraft users group on line?  Fortunately I have my brand new KX3, so I can concentrate on POTA for the next several months.

 

73 de Gil K4JST     birnamwood@widomaker.com

[K3MRI]

@K3MZ @W4TG Any ideas Paul or Frank?

[KN3U]

Gil,

There is an Elecraft forum at elecraft@groups.io. I am not a frequent visitor, but the forum has a number of very knowledgeable members. It is highly likely that you will find recommendations for repair services there, and probably also troubleshooting tips.

It seems likely that the radio experienced a voltage surge. Diode D28 is a solid-state transient voltage suppressor (TVS). It is similar to a Zener diode, but designed to absorb moderately high-energy surges. it presents a low resistance to voltage exceeding a specified threshold, clamping the voltage seen by the downstream circuitry. If the overvoltage condition is extreme, the current flowing through the TVS will blow the radio's main fuse. WHen a TVS fails, it usually fails in a short-circuit condition, sacrificing itself to protect the downstream circuitry.

That particular part number, BZW50-15, starts conducting when the incoming voltage reaches between 18 and 20 VDC. Because the TVS has a small but measurable resistance, if the surge current is large enough -- for example, due to a nearby lightning strike -- the clamping voltage seen by the downstream circuitry may rise to as much as 30 or 40 V for a brief instant before the fuse blows. That's enough to cause downstream damage, and you might have multiple failures.

Your symptoms make it sound like the power supply voltage coming into the radio did experience a voltage surge. If you were not experiencing lightning conditions during Field Day, I would be very suspicious of your power supply. In many power supply designs, the failure of a series pass transistor can apply a large unregulated DC voltage to any connected equipment. In addition, if the power supply  has a voltage adjustment, the potentiometer that controls the output voltage may become intermittent with age and result in a transient overvoltage condition. So this is something to look at, lest you damage another radio.

Better-quality analog power supplies have an independent protection circuit called a crowbar circuit to prevent an overvoltage condition. Switching power supplies may employ a variety of overvoltage protection schemes having varying effectiveness, or none at all. Not all power supplies sold in the ham radio market -- including some major brands -- have overvoltage protection.

Good luck.

Al KN3U

[K4JST]

Al KN3U,

Thanks for the input and explanation of just what D28 does.  I did contact Elecraft, but have not heard from their tech yet.

The rig was plugged in to my 100 AH LiPO4 battery (I use this all the time at home as the only power source) and had been operating for perhaps 14 hours at that time, so power supply surge was not the problem.  I was away from the rig and another club member decided to switch batteries to a solar charged one.  Same power cord and both batteries had correctly wired Power Pole connectors so reverse polarity is not suspected.  They did smell smoke and the rig was dead.  Since then, I replaced D28.  The rig comes up on 40 cw, but audio with no signals. Rig will not tune to another frequency and can not bring up the menu.  Tried the factory reset and get EE INIT for a few seconds, but no error code followed by the unit locking up and display goes blank.

I suspect multiple areas are damaged and have little hope that I can track them down and repair.  The P3 pan which was connected appears to be un damaged, but will check that with a working K3 before I know for sure.

Thanks for your input,

de Gil K4JST

[K4JST]

To Al KN3U or others:

What is the name of the K3S configuration file?  I know I saved it when I got the rig, but am faced with finding it on any of three computers and perhaps an old hard rive or two.  Without a name or format for the name, I am getting now where finding it again.  NR4C Bill the original owner is a SK and his computer is long gone.

 

de Gil K4JST

[K3MRI]

@KN3U @W4TG @K3MZ Can you guys help? I have no idea...

 

36 minutes ago, K4JST said:

To Al KN3U or others: What is the name of the K3S configuration file?  I know I saved it when I got the rig, but am faced with finding it on any of three computers and perhaps an old hard rive or two.  Without a name or format for the name, I am getting now where finding it again.  NR4C Bill the original owner is a SK and his computer is long gone. de Gil K4JST

 

[KN3U]

Gil,

Look here:

C:\Users\<your username>\AppData\Roaming\Elecraft\K3 Configuration\SN06398.20220731T143229.K3Config

The filename has the serial number, period, date (YYYYMMDD), the letter T, and the time (HHMMSS) the file was created.

Or just search for "*.K3config".

Al KN3U

[K4JST]

Al KN3U,

Thanks for the help with the configuration file.

So far no luck with finding the .K3config file on any of the computers or HDs I might have had at the time.  I did try *10062* the serial number of the K3S.  It is surprising the number of files the Search returned with those numbers in them (old photo files names etc), but still no luck.  Waiting to hear from Elecraft regarding sending it back for repair.  They say there is a real back up for repairs right now.  I do have my KX3 to play with and borrowed the clubs K3 to do some A-B checks, but that is just plain frustrating! 

73 de Gil K4JST

[WA2WMR]

Can you get to "C:\Users\<your username>\AppData\Roaming\Elecraft\K3 Configuration\"?

[K4JST]

Lind WA2WMR,

 

Yes I have searched that location on all three of the computers that I had at the time.  No luck.  Also searched the old hard drives that I have lying around using

SN10062.*.K3config 

No luck with that either.  One more chance, another club member may have Bill's NR4C s computer or his hard drive or files (NR4C is a SK).

I did down load the Config file from my new KX3 and the same file from a K3 that NR4C owned.  Unfortunately,  I can not get a connection to my K3S.  Someone suggested using the Tool 's menu that is part of HRD, but I suspect it will not connect either.

 

The K3S comes up and I can now change frequency, but can not get to the Display and the rig will still not powerdown until the power cord is disconnected from the battery.  

Strongly suspect that this is going to go back to Elecraft for repair, but they will not even send me an RA yet since they are far behind with repairs. Gives me a good excuse to play with my new KX3 (we actually used it for several hours at FD until another rig was located).

My thanks to you and Al and all on Hamcommunity for your input

73 de Gil K4JST

 

[KN3U]

One other thought -- you might try reloading the firmware using the K3 Utility program. Chances are that you have already tried this.

First, try a factory reset. In the manual (page72), it is called Parameter Initialization. Then see if you can connect to the radio using the K3 utility. If so, you can try sending all the firmware to the radio.

[K4JST]

Al,

Still trying to locate a copy of the K3S configuration file before I try a factory reset.  A couple of club members may have NR4C s computers or file backups.  
 

73 de Gil K4JST

[KN3U]

I understand. The factory reset does install a default configuration file, so if the reset is successful, the radio will be fully functional. You can always install a custom config file later, if one is found. There is plenty of advice to be found on the web regarding ways to customize the configuration for different pursuits such as DXing, contesting, or VHF/UHF activities. Fred Cady, K37X, wrote a book on the K3 with more detail than the manual, including advice on how to set up the radio for different pursuits. Elecraft sells the book, and I think Amazon carries it as well.

Talk to you soon,

Al KN3U

[K4JST]

Actually, I have a copy of his book.  Need to take a look at that section and get away from my fear of creating a Config file.  I just remember Bill (NR4C) telling me that it was not for the faint at heart.

Will be at my sister’s this weekend for my 60th HS reunion (where does the time go?).  Will give you a call and perhaps we can visit?  
 

some how, I just got the K3 Utility program to connect to the K3S.  Down loaded the Config file and saved it on the computer.  Then updated the time and date on the K3S, so the computer is talking to the rig again!  Have not been able to do the Factory reset so far though.  Getting late, so will spend some more time in the morning.

Thanks for your help and will keep you updated on the rest of this mess.

73 de Gil

[K4JST]

To all who are following K3S damage:

Have been back into the rig. The large diode on the 100 w amp board is not damaged ( took an hour to get to board, test and reinstall it).  Makes me appreciate all who built a K3 or K3S from the “kit”. Both the 5 volt and 8 volt regulators are working correctly.  Need to go in and test Q2.  Since this rig has the sub receiver installed, this is another hour project.  Sure am learning about how my rig is put together! 

 

Thanks again to all who have given me input on this.  I am impressed with Ham Community

 

73 de Gil K4JST

[K3MRI]

2 hours ago, K4JST said:

Thanks again to all who have given me input on this.  I am impressed with Ham Community

73 de Gil K4JST

We're new; we're small; but we care 🥰

[K4JST]

Actually, getting to the Q2 is really easy.  Simply remove the bottom panel of the K3S.  The CMOSFET is SMT on the bottom of the board. Digikey has the part.  Does anyone know how to test the part for failure without removing it?  I really have had little experience removing SMT parts and replacing them (and the RF board is $600)!  Certainly getting to the edge of my comfort zone.

 

73 de Gil K4JST

[KN3U]

Q2 on the RF board is a P-channel MOSFET functioning as a power switch. You can probe it in place. The source (pins 1, 2, 3) are connected to the 12 VDC main power input through a polyswitch (resettable fuse) and a Schottky diode. Verify that you have 12 V on Pins 1-3. Pin 4 is the gate of Q2. When it is grounded, the MOSFET turns on and 12 V should appear on the drain (pin 5-8).

The signal that grounds the gate of Q2 comes from the front panel PCB. It might be hard to probe Pin 4 directly, but you can put your probe on any of the components connected to pin 4 -- R76, R77, or C226.

If Pin 4 doesn't go to ground when you turn on the radio, you'll have to go back further to the front panel PCB. On that board, there is a signal called /PWRON that is driven by the Microprocessor. It's unclear from the schematic exactly how that works, but here's what I think might be happening.

It appears that the Power Switch is a momentary pushbutton. When you press the power switch, that turns on Q3, which grounds the /PWRON line (and goes to the gate of Q2 on the RF board. That applies power to the radio, including the microprocessor. The microprocessor then asserts the POWERHOLDON signal, which turns on Q2 on the front panel PCB. That keeps the /PWRON line grounded when you release the power button. A second push of the power switch sends a different signal to the microprocessor to shut the radio down.

So, the bottom line is this: In order for the radio to turn on and stay on, Q2, Q3, and the microprocessor on the front panel PCB all have to work, as well as Q2 on the RF board. Looking at the gate of Q2 on the RF board, press and hold the power switch and see if the gate goes to ground. If so, power will come on on the RF board. Then see if the gate remains grounded when you release the power switch.

My guess, since you have been able to communicate with the microprocessor, is that the circuitry on the front panel PCB is ok and your issue is a bad Q2 on the RF board.

[K4JST]

Al KN3U,

I just closed up the rig after attempting to trouble shoot Q2 the P channel MOSFET.  It tests ok, but I need to go back and follow along with your latest.  When I was doing the testing, I noticed audio coming through the speaker.  Connected a short ant to ANT 1 and then brought my KX3 to that frequency and ran 1 watt to the outside antenna.  Sounded like perfect CW when I keyed the KX3.  Tried it for a half hour and just fine.  I can not change the freq of the K3S.  It always come up with the frequency locked and when I press and hold Unlock, all audio goes away, but I can change the frequency.  If I try and power the K3S off, the display goes blank ( backlight stays on) and I have to unplug the power cord to do anything else.

Just getting ready to figure out how to get to the front panel, but will check several things you suggested ie R76,R77. And C226.

Will be later this evening, company coming for burgers.

 

Thanks again for all your help,

73 de Gil

[WB3GBL]

Hi Gil:  I know you guys are knee deep into troubleshooting the RF board and the front panel circuits, but I keep thinking back to field day and some of the odd things that occurred around the time the K3S went bad. The laptop went out and that was connected by way of N1MM and those signals travel thru the KIO3B board. Some people have reported large currents / voltages causing damage thru the communications ports as well as symptoms exactly like yours with KIO3B problems. Just wondering if its possible to run the rig without the KIO3B and the daughter boards?  I know you have established communications thru these ports to your rig / computer - so maybe I'm just barking up the wrong antenna tower hi hi. But anyway just letting you know I'm rooting for you guys!  73  wb3gbl rick

[K4JST]

Rick,

Will look at removing the KIO3B board.  And thanks for taking time to follow this.

For others following this, Rick is a member of WAARC and was at FD and watched the tracking down of the D28 which was burned and cracked.  Had no idea he was following this.

73 de Gil K4JST

looks like a number of members of WAARC (Williamsburg Area ARC) are following the forum!

[K4JST]

Rick WB3GBL and others,

 

I removed the KIOB3 board.  There was no change  ie rig comes up, when I unlock to allow changing freq display freezes and Ican do nothing else..

This rig has the sub receiver and I am getting much better at removing it and replacing it

 

Thanks to Rick WB3GBL for suggesting this.

de Gil K4JST

[WB3GBL]

Hi Gil:  Ah, disappointing.  But I believe we're all looking at this thing as a TTL logic problem and not some out of tolerance component.  You'll probably go ahead and dig into the front panel circuit.  Since you'll already be disassembling the front panel to get at the Q2, Q3 possibilities - the most cited component that brings on these type of faults is C102 (10uf, 10V electrolytic cap). TTL levels here could be +5 or +3.3 volts, without these voltages available for the PWR signals the radio won't respond properly - this is where a shorted C102 can ground out the TTL voltages and screw things up for needed logic. Thanks for continuing into the Elecraft K3S problem as the comments here can help K3S owners years from now.  Good luck and hope KN3U can respond with more insight after your findings.  73 wb3gbl rick

[KN3U]

These symptoms that you are describing relate to functions implemented in software, not hardware. All the front panel switches and rotary encoders are read as DC voltages by an analog-to-digital converter (ADC) built into the the microprocessor. See page 4 of the front panel schematic. A voltage divider string connected between 5 VDC and ground applies a different DC voltage to each string of switches. The voltage divider has seven "taps" ranging from 1.5 V to 4.5 V in half-volt increments. There are eight inputs to the ADC. The microprocessor constantly scans the ADC inputs and knows which switch is being pressed by the voltage level and which ADC input it appears on. For example, the tuning rate switch is connected between TAP 6 on the voltage divider and ADC input 1. TAP 6 is 2 VDC. So when the microprocessor reads 2 VDC on ADC input 1, it knows the rate switch is pressed. (I'm simplifying things somewhat, but that's the general idea.)

The bottom line is that, if the 5-volt regulator is malfunctioning and putting out a voltage that is different from 5 VDC, the microprocessor will be confused as to which switch is being pressed. So it is worth checking that supply voltage. The 5-volt regulator on the front panel PCB is U10, a 78M05 chip.

Rick, C102 on the front panel board filters the contrast voltage applied to the LCD display, so that's not likely to be the culprit here. I'm curious, though. It sounds like you have found a source of troubleshooting information. If so, I'd like to know about it. I'm just working from the K3S schematic diagrams published by Elecraft.

Ak KN3U

[K4JST]

5 hours ago, KN3U said:

These symptoms that you are describing relate to functions implemented in software, not hardware. All the front panel switches and rotary encoders are read as DC voltages by an analog-to-digital converter (ADC) built into the the microprocessor. See page 4 of the front panel schematic. A voltage divider string connected between 5 VDC and ground applies a different DC voltage to each string of switches. The voltage divider has seven "taps" ranging from 1.5 V to 4.5 V in half-volt increments. There are eight inputs to the ADC. The microprocessor constantly scans the ADC inputs and knows which switch is being pressed by the voltage level and which ADC input it appears on. For example, the tuning rate switch is connected between TAP 6 on the voltage divider and ADC input 1. TAP 6 is 2 VDC. So when the microprocessor reads 2 VDC on ADC input 1, it knows the rate switch is pressed. (I'm simplifying things somewhat, but that's the general idea.)

The bottom line is that, if the 5-volt regulator is malfunctioning and putting out a voltage that is different from 5 VDC, the microprocessor will be confused as to which switch is being pressed. So it is worth checking that supply voltage. The 5-volt regulator on the front panel PCB is U10, a 78M05 chip.

Rick, C102 on the front panel board filters the contrast voltage applied to the LCD display, so that's not likely to be the culprit here. I'm curious, though. It sounds like you have found a source of troubleshooting information. If so, I'd like to know about it. I'm just working from the K3S schematic diagrams published by Elecraft.

Ak KN3U